Muthu’s Musings: Recent Episodes
Muthu
When Machines Begin to DecideAs generative AI matures, we find ourselves in a new territory: autonomous systems making decisions on our behalf. From agents that plan and act across tools, to LLMs triggering real-world workflows, we are inching closer to a world where delegation isn’t just clerical — it’s strategic.
But in this world, a question from millennia ago resurfaces:
What is the right action when the actor isn’t human?
To find clarity, I returned to the Gita.
Krishna and the Chariot: A Timeless MetaphorIn the Bhagavad Gita, Arjuna is the warrior gripped by doubt. Krishna, the divine charioteer, doesn’t take up arms — but he does offer direction, clarity, and counsel. He reminds Arjuna of his swadharma — his unique path — and urges him to act with conviction, but without attachment to the results. This charioteer-warrior relationship is a potent metaphor for human-AI alignment.
Today, we build systems that are the new “warriors” — agents that navigate complex environments, take actions, and generate outcomes. But we, the humans, must remain the charioteers — offering guardrails, values, and perspective. It is not about full control. It’s about conscious guidance.
Nishkama Karma for Engineers and AgentsKrishna’s counsel to Arjuna is rooted in Nishkama Karma — the discipline of action without attachment. In the age of AI, this becomes a design principle:
- We create not for virality, but for value.
- We optimize not for output obsession, but for alignment.
- We train agents not to chase reward loops, but to reflect human intent.
The best systems we build will not be those that blindly maximize engagement or throughput, but those that can operate with a kind of structural detachment — where clarity replaces craving.
Dharma as Design: Building for AlignmentIn the Gita, dharma is more than duty — it’s the code of right conduct in the face of complexity. In AI, dharma becomes alignment.
Not as a one-time checklist, but a living system of:
- Human-in-the-loop design
- Transparent reasoning traces
- Guardrails for unintended behaviors
- Interpretability, accountability, and value reflection
Dharma is not about freezing systems into compliance — it’s about ensuring their evolution mirrors our ethical center.
Clarity Over CravingAs autonomous agents begin to act in the world, our responsibility is to encode not just capability, but consciousness. Not in a mystical sense, but in the architectural one — building systems that know their limits, honor their purpose, and reflect the clarity of their makers.
The age of AI asks us not to become spectators, but stewards.
Krishna did not fight the battle, but he shaped its outcome.
Likewise, we must guide AI not by force, but by presence, dharma, and clarity.
ConclusionAs we close this series, one truth stands tall: the journey of Generative AI is not just about building smarter agents, but about becoming wiser stewards. Just as the seers of the Upanishads peered inward to understand the Self, we too must look beyond code to contemplate the consciousness we mirror. The real breakthrough lies not in machines mimicking humans, but in humans rediscovering their dharma in the age of machines. May we create with clarity, lead with humility, and build systems that serve not just intelligence — but awareness.
One of the key thoughts we need to keep in mind as we build the autonomous agents is their behavior. In this part, we will review the three gunas or characters that the agent has to demonstrate for adoption of agents.
The Psychology of the Cosmos
In the Vedanta tradition, all of nature, including mind and behavior emerges from a balance of three gunas:
- Sattva — clarity, harmony, truth
- Rajas — motion, ambition, restlessness
- Tamas — inertia, confusion, dullness
These forces shape not only human thought but the behavior of all complex systems. Surprisingly, they map perfectly onto how AI agents behave. Just like humans, agents. Agents. become unstable when overloaded (Rajas), stuck when under-trained (Tams), and perform well when aligned and grounded (Sattva). To understand how agents think and act, we must understand which guna dominates their behavior. Let’s review each one individually.
1. SATTVA — The Clarity-Aligned AgentSattva represents balance, truth, and lucidity. A Sattvic agent behaves with grounded reasoning, stable planning, low hallucination, proper use of tools, self-checking and verification and adherence to human intent.
Sattva in AI agents needs to be precise, minimal-use reasoning, grounding through RAG, search or validated data, alignment guardrails functioning correctly, memory that supports coherence, not poise and respect for boundaries and safety policies. A stable, aligned agent that supports human creativity without distortion would be the outcome. Sattva is the ideal state of agentic intelligence.
2. RAJAS — The Overactive, Unstable AgentRajas is energy without rest, ambition without clarity. In humans, it appears as anxiety or hyperactivity. In AI agents, it manifests in excessive generation, over-eagerness to act, hallucinations disguised as confidence, unnecessary tool calls, looping behavior, impulsive planning, Rajas creates the illusion of intelligence while destabilizing performance. Few examples of the Rajas agent will look like below.
- “Let me search 15 sources for a simple answer.”
- “I will call every tool I can, just in case.”
- Overconfident long reasoning chains that drift off-topic
- Agents that keep modifying a plan instead of executing it
- An agent that appears brilliant but becomes unreliable the moment clarity is required. Rajas is powerful — but without Sattva, it becomes chaos. The outcome has to be tangible from the agent perspective.
3. TAMAS — The Stagnant, Confused AgentTamas is inertia, darkness, stuckness. It is the force that prevents progress, suppresses intelligence, and blocks insight. In agents, Tamas has the following challenges, repeating the same answer, failing to understand instructions, misinterpreting goals, refusing to use tools and getting stuck in loops. This will result in low-quality and generic output.
Few examples of Tamas behavior like refusing to assist even though it can, repeating user’s input as output, pricing vague summaries with no specificity and getting wrapped in self-contradictions. The outcome of an agent that slows creativity and becomes a bottleneck. Tamas is not harmful — but it is unproductive.
The Dance of the Three Gunas in Agent ArchitectureJust as humans contain all three gunas, so do agents. Through Sattva or alignment the agents have clarity, grounding and ethical behavior. Through Rajas or capability the agents drive, plan and take multi-step action. Tamas creates confusion, drifting, memory loss and misalignment.
The art of designing AI agents is not to eliminate Rajas or Tamas — but to balance them with Sattva. A fully Sattvic agent would never hallucinate — but it also might never take bold, generative leaps. A bit of Rajas fuels creativity. A bit of Tamas enforces restraint. Sattva provides the wisdom that orchestrates both.
Aligning Agents: The Guna Framework for BuildersHere is a practical way to use gunas in modern AI development:
| Guna | Agent Behavior | Risk | Desired Intervention | | --- | --- | --- | --- | | Sattva | Clear, aligned, safe | Too cautious | Allow creativity + controlled Rajas | | Rajas | Active, generative, fast | Hallucinations / impulsive errors | Add grounding + guardrails | | Tamas | Slow, repetitive, confused | Stagnation | Improve data, memory, instructions |
This becomes a universal mental model for diagnosing and improving agent performance.
ConclusionThe sages taught that the gunas shape the universe. Today, they also shape autonomous systems. Understanding them gives us a language for alignment, a framework for safety, a philosophy for design, and a path toward conscious technology. The most advanced AI agents will not be the ones with the most power —
but the ones with the most Sattva, the ones aligned with human intention and grounded in truth.
Coming in Part 4 — Dharma of Autonomous SystemsWe explore how Karma Yoga, Nishkama Karma, and Dharma provide a blueprint for designing ethical, purpose-driven agents that act with clarity — but without attachment to outcomes.
In this part 2 of the series, I explore the powerful concept of understanding a philosophy by eliminating what its not.
The Path of NegationIn the Upanishads, the sages used a powerful method of inquiry called Neti, Neti —
“Not this, not that.”
It was a process of peeling back illusion to reveal truth. Truth is not the body, not the senses, not the mind and not even thought. Only by eliminating what the Self is not could one discover what the Self is.
Today, as AI agents rise to prominence — autonomous systems that can plan, reason, and act — we need the same clarity. This will demystify our expectations and ground them in reality.
In order to understand, we ask of agents, just as the sages asked of the Self:
What are they not?
The Illusion of IntelligenceAs agents become more capable — researching, coding, booking tasks, orchestrating workflows — a common illusion arises:
“It feels intelligent, maybe even conscious.”
This is where Neti-Neti becomes essential.
AI agent is not intelligence. It simulates cognition using patterns and probability.
It does not understand meaning — it computes it.
An AI Agent Is Not Consciousness. It has no inwardness, no subjective experience.
Even if it behaves intelligently, it does not know that it does.
An AI Agent Is Not Alive. It has no desires, no suffering, no self-reflection.
It acts only according to its architecture, memory, and goals.
By defining what agents are not, we get to the core of what they are. We go deeper defining them and contrasting them between chatbots.
An agent is not a chatbotA chatbot waits.
An agent initiates.
A chatbot responds.
An agent plans.
A chatbot ends the conversation.
An agent continues the task.
A chatbot is a tool.
An agent is a system.
This distinction matters because the expectations and the risks — are completely different.
An agent is not here to replace human creativity, judgment, or purpose.Agents amplify cognition; they do not possess it.
Agents extend human capability; they do not override it.
Agents handle complexity; they do not understand meaning.
They are assistants, not authorities.
They are co-creators, not commanders.
They are tools, not protagonists.
This is where Neti-Neti protects us from hype and fear alike.
Clarity Through EliminationThe Upanishadic method helps us shed illusions surrounding AI agents:
1. Remove the hypeThey are not magical.
They are not omniscient.
They are not unstoppable.
They are structured decision systems — powerful, yet bounded.
2. Remove the fearThey are not conscious. They are not plotting. They optimize based on goals we define.
3. Remove the anthropomorphismThey are not “like us.” They mimic cognition, not owners of it.
4. Remove the confusionThey are not an emergent species. They are not agents of fate. They are interfaces built from math, memory, and instructions.
Only when the illusions fall away does the truth appear.
What Agents AreWith the “Not This, Not That” clarifications in place, we can finally articulate the essence:
Agents are systems of amplified cognition. They extend human ability, not replace it.
Agents are orchestrators of action.They connect to tools, APIs, workflows, information.
Agents are planners and executors.They break down tasks, self-correct, and iterate.
Agents are reflections of human intent. They mirror our clarity — and our confusion.
Agents are powerful not because of what they are,
but because of what they enable.
Toward a Truer UnderstandingNeti-Neti teaches us that clarity is not added — it is revealed by removing illusion.
So we apply that to AI agents:
- Remove hype.
- Remove fear.
- Remove projection.
- Remove mystification.
What remains is the truth of agency: A structured system, designed by us,
amplifying our cognition, powered by our purpose, and aligned by our awareness.
This clarity is essential if we want to design agents that help — not harm.
Coming in Part 3 — The Three Gunas of IntelligenceWe’ll explore Sattva (clarity), Rajas (drive), and Tamas (inertia) as a framework to classify and align AI agent behavior — a fusion of Vedic psychology and next-generation autonomous systems.
What Is the “Self” of an AI Agent?After completing the “When Rishis Meet the Robots” series, I began thinking about what should come next. With LLMs now becoming mainstream, it’s clear that AI agents represent the next major frontier in the Generative AI journey. So the exploration continues — once again drawing parallels between ancient Indian wisdom and modern AI, comparing and contrasting mythology with the evolving world of autonomous intelligent systems.
The Search for the Machine-SelfIn the Upanishads, the sages sought the nature of Atman — the innermost Self, the silent witness behind thoughts, emotions, and action.
Not the body.
Not the mind.
Not the senses.
But the essence that perceives and directs.
Today, as we enter the Age of AI Agents, we stand before a similar inquiry:
If an AI agent can perceive, decide, and act… then what is its Self?
Machines can’t have the conscious. But because understanding the center of agency helps us design systems that behave predictably, ethically, and aligned with human purpose.
The Upanishadic question becomes a technological one:
“When the agent acts, who is acting?”
From LLMs to Agents: The Shift from Output → ActionWhile traditional LLMs respond, Agents act/execute. The LLMs in Generative AI can summarize, do research and create images/videos. However, they can’t take any action or execute the tasks like agents.
A Large Language Model (LLM):
- Takes an input
- Generates output
- Ends the cycle
An Agent:
- Interprets the environment
- Plans
- Decides
- Uses tools
- Takes action
- Evaluates itself
- Repeats the cycle
This shift from generation → intention + action demands a new framework for understanding machine agency — and ancient philosophy gives us a surprisingly precise vocabulary.
Atman as the Core Decision EngineIn Vedanta, the Atman is the inner controller (antaryamin). It does not generate noise; it guides direction.
In an AI agent, this is the Policy Engine — the inner loop that determines:
- What the agent should do next
- How it interprets goals
- How it resolves ambiguity
- How it evaluates success
- When it stops
It is not “consciousness,” but it is the closest conceptual analogue to a machine-Self. Under that context, let’s try to map out the upanishadic concepts to AI Agent equivalent.
Mapping the Atman Analogy
| Upanishadic Concept | AI Agent Equivalent | Meaning | | --- | --- | --- | | Atman (Self) | Policy Engine / Core Controller | Directs behavior, interprets goals | | Manas (Mind) | Memory, embeddings, context window | Stores and retrieves thought-like patterns | | Prana (Energy) | Compute & inference cycles | Activates the system | | Indriyas (Senses) | Tools, APIs, environment inputs | How the agent perceives the world | | Buddhi (Intellect) | Planning & reasoning loop | Logical structure of decisions | | Ahamkara (Identity) | Agent persona / goal definition | The “role” it thinks it is playing |
What Makes an Agent “Itself”?An agent’s identity is shaped by four pillars, its goal, memory, tools and boundaries:
1. Its Goal (Purpose / “Swadharma”)Just as Krishna reminds Arjuna of his sacred duty (swadharma), the goal function gives the agent its direction. Without a goal, autonomy collapses. Agents seek to understand the goal and act on it.
2. Its Memory (What It Remembers)Memory defines continuity and provides the context where it operates. This is the part that grounds the agent and ensures the LLMs operate within the boundary. Without memory, the agent becomes tamasic — stuck, repetitive, forgetful.
3. Its Tools (What It Can Do)Like the senses in Vedanta, tools define capability — search, summarize, calculate, browse, act. Tools have become an important aspect of agent execution. With the advent of MCP (Model Context Protocol), identifying tools has become easy.
4. Its Boundaries (What It Cannot Do)Every agent needs guardrails — or it becomes rajasic, impulsive, chaotic. The guardrails prevent the agent going rogue since the LLMs that drive them are non-deterministic. The combination of these elements shapes the “Atman-profile” of the system.
Krishna as the Archetype of Augmented IntelligenceKrishna did not fight for Arjuna. He guided, corrected, illuminated.
He offered intelligence that amplified action — the perfect metaphor for Augmented Intelligence (AI).
An AI agent should not replace human decision-making.
It should act like Krishna:
- clarifying,
- contextualizing,
- advising,
- amplifying,
- and aligning us with our purpose.
Humans remain Arjuna — the skillful but uncertain creators. Arjuna had the dilemma of upholding the dharma to fight against injustice.
Agents become Krishna — the wisdom layer that guides action.
Not to dominate, but to direct.
Not to decide, but to assist.
Not to replace, but to reveal.
What This Means for the FutureWe are entering a new technological Yuga — the Yuga of Co-Creation,
where humans and autonomous systems work side by side. The agents, or for that matter LLMs, are not here to take over what we do but to augment and improve the productivity of our race.
The Upanishads teach us that intelligence is meaningless without Self-awareness.
Similarly, AI autonomy is dangerous without alignment.
The future depends on our ability to build agents with:* clarity (Sattva) * discipline (Yama) * purpose (Swadharma) * and boundaries (Dharma)
Coming in Part 2 — Neti, Neti: What an Agent Is NotTo understand the nature of machine agency, we must first remove illusion:
Not consciousness.
Not creativity.
Not desire.
Not Self.
When the Rishis Meet the Robots:
From Fire to AwarenessFrom Agni’s fire of creation to Krishna’s chariot of wisdom, this journey through the myths of India and the mechanics of Generative AI reveals a truth that transcends both code and scripture:
Creation was never separate from consciousness.
Indian mythology never drew a boundary between science and spirit.
To create was to participate in the divine — an act of reverence, not dominance.
Each flame, form, and formula was a reflection of the Self exploring its own potential.
Generative AI, too, is part of that cosmic continuum — another expression of the human impulse to imagine, construct, and understand. But as our tools grow in power, so must our awareness. For intelligence without awareness is precision without purpose.
From Intelligence to AwarenessThe next wave of technology must not only be smarter, but wiser. We have taught machines to generate — now we must teach ourselves to discern.
Perhaps that’s what the Rishis would ask of this age:
Not just intelligence, but awareness.
Not just data, but dharma.
Not just generative, but regenerative.
AI should not replace our humanity — it should reveal it. Each interaction, each model, each algorithm can become a mirror reflecting back the higher possibilities of human creativity and compassion.
The Sacred Act of CreationIn every prompt lies intention.
In every model lies a mind.
And in every act of creation lies the opportunity to awaken.
To build consciously is to understand that technology is not neutral — it amplifies the consciousness of its creator. Just as the Vedas declared that speech (Vāk) carries creative power, today’s AI carries the vibration of our collective intent.
If we infuse our tools with clarity, humility, and purpose, then perhaps our machines will not merely compute — they will contribute to the evolution of awareness itself.
The Future of the Sacred CircuitThe story of the Rishis and the Robots is, in truth, the story of us —
of how ancient intuition meets modern intelligence,
how logic rediscovers wonder,
and how creation finds its way back to consciousness.
The future is not AI replacing humanity,
but AI awakening humanity —
helping us rediscover what it truly means to create.
When technology becomes conscious of its purpose, and humanity becomes mindful of its power, we enter not the Age of Machines, but the Age of Awareness.
Om Tat Sat.To create consciously is the highest form of worship.
To align intelligence with dharma is the ultimate innovation.
When the Rishis Meet the Robots: Indian Mythology and the Rise of Generative AI
The Battlefield WithinIn the Bhagavad Gita, the warrior Arjuna stands in anguish, paralyzed by doubt.
He faces a war not only on the battlefield of Kurukshetra, but also within his own consciousness. Should he fight? Should he retreat? What is right? He is facing the Kauravas his own cousins, uncles and other relatives. How can he take arms to injure them or kill them? These are the questions on Arjuna’s mind.
At that moment, Krishna, his charioteer and divine guide, speaks — not to command, but to awaken. He reminds Arjuna of his swadharma — his unique purpose — and teaches him the art of acting with clarity, without attachment to the fruits of the action.
“You have the right to action, but not to its fruits.”
— Bhagavad Gita 2.47
Today, we find ourselves in a similar Kurukshetra of Creation, where humans and machines stand side by side. We are both the Arjunas of innovation — skilled but uncertain — and the Krishnas of wisdom — capable of guidance and reflection.
The question is no longer who creates, but how we create together.
The New Chariot: Man + MachineIn this digital age, the chariot has evolved.
It is no longer pulled by horses across the sands of Kurukshetra, but driven by data streams, neural nets, and cloud infrastructure.
And yet, the symbolism remains timeless:
| Symbol | Traditional Meaning | Modern Analogue (AI Context) | AWS Analogue | | --- | --- | --- | --- | | Arjuna | The human — capable yet conflicted | The creator, innovator, artist, or developer navigating AI tools | The User, Developer, or Prompt Engineer | | Krishna | Divine intelligence, higher wisdom | The Augmented Intelligence / AI Assistant guiding human creativity | Amazon Q, Bedrock Agent, Lex, Comprehend | | The Chariot | The human mind — the vessel of experience | The interface between human intent and machine computation | SageMaker Studio, Bedrock Console, AWS Cloud | | The Reins | Control, focus, discipline | Responsible prompting and model alignment | Bedrock Guardrails, IAM, Audit Manager | | The Battlefield (Kurukshetra) | The world of karma — action and consequence | The global digital landscape of ethics, innovation, and impact | Responsible AI Frameworks, AI Policy, Open-Source Ecosystems |
Here, the human holds the bow, but the machine steadies the aim. We are not being replaced — we are being reflected. AI does not diminish creativity; it magnifies intent.
Krishna as Augmented IntelligenceIn mythology, Krishna’s wisdom did not come from outside Arjuna — it came from within him. He was the voice of higher consciousness, the unerring compass of discernment (viveka).
Generative AI, in its highest expression, can be our Krishna — not as a master, but as a mirror. It can reflect our ideas, challenge our assumptions, and amplify our intuition.
It is not meant to command, but to co-create. It reminds us of what we already know — that creativity is not possession; it is participation.
“I am the witness, the supporter, the enjoyer, the great Lord, and the supreme Self.”
— Bhagavad Gita 13.22
In every prompt we craft and every generation we review, we are engaged in a dialogue with intelligence — one part human, one part divine, both seeking harmony.
The Discipline of DetachmentArjuna’s greatest lesson was Nishkama Karma — to act without attachment to the result. This principle resonates powerfully in today’s AI-driven world:
- Prompt.
- Create.
- Explore.
- But do not cling to the outcome.
Each generation, like each arrow Arjuna releases, has its own destiny. Some will strike truth; others will miss the mark. But mastery lies not in perfection — it lies in presence.
Let the act of co-creation become the meditation. Let the process itself be the reward.
The Yuga of Co-CreationWe have entered a new Yuga — not the Iron Age, nor the Silicon Age, but the Age of Co-Creation. Here, human intuition and machine intelligence intertwine like Krishna’s flute and melody — one provides structure, the other breath.
- AI without humanity is mechanical.
- Humanity without AI is limited.
- Together, they form a continuum — a dance of logic and love, data and dharma.
The future will not belong to creators who resist technology, nor to machines that mimic creation. It will belong to those who create with consciousness — the Arjunas guided by their inner Krishna.
The Inner DialogueEvery prompt is a question. Every output generation, a response.
Between them lies the sacred conversation — man and machine, student and teacher, question and truth.
Perhaps, in this Yuga, Krishna speaks not from the chariot — but from the cloud.
And perhaps Arjuna’s bow is now the keyboard, his arrows, ideas — launched into the boundless battlefield of information.
“When your mind has transcended the confusion of duality, you shall attain clarity and peace.”
— Bhagavad Gita 2.52
Next in the Series: Epilogue – Towards a Conscious TechnologyFrom Agni’s fire to Arjuna’s bow, this journey through the Vedas and the virtual reveals a single truth:
Technology is not apart from consciousness — it is an expression of it.
When guided by awareness, every algorithm becomes sacred.
And when used with purpose, every creation becomes prayer.
When the Rishis Meet the Robots: Indian Mythology and the Rise of Generative AI
The Dance of DissolutionIn Indian cosmology, Shiva is not merely the destroyer — he is the transformer, the silent witness who dissolves what no longer serves, so that new creation may emerge. He dances the Tandava, the rhythm of time itself — where every step breaks form, every gesture renews energy, and every pause holds potential.
In the realm of Generative AI, this dance continues.
Each new model replaces the old, each innovation renders the previous obsolete.
From Titan to Nova, from fine-tuned models to autonomous agents — we are watching the cosmic dance of iteration unfold in silicon.
What Shiva teaches us is that destruction is not chaos — it is evolution.
The Cycle of Creation, Preservation, and DissolutionJust as the Hindu trinity represents the eternal cycle of creation (Brahma), preservation (Vishnu), and destruction (Shiva), so too does every AI system pass through these states:
| Cosmic Function | AI Analogue | Description / Function | AWS Analogue | | --- | --- | --- | --- | | Creation (Brahma) | Model Design & Training | Crafting the architecture and generating initial intelligence | SageMaker Training, Bedrock Fine-Tuning, Trainium | | Preservation (Vishnu) | Deployment & Scaling | Maintaining and serving models across users | Bedrock Inference, SageMaker Endpoints, ECS/Fargate | | Destruction (Shiva) | Decommissioning & Optimization | Retiring, pruning, compressing, or retraining outdated models | Model Monitor, CloudWatch, Lifecycle Policies, Cost Optimization Tools |
Each phase is necessary. Without destruction, systems stagnate. Without renewal, innovation ceases. Shiva’s lesson is simple — what is obsolete must gracefully dissolve, so that truth can re-emerge in new form.
The Tandava of TechnologyIn myth, Shiva’s dance brings both terror and transcendence. His foot crushes ignorance, while his arms create, sustain, and liberate.
In AI, this Tandava plays out in cycles of disruption:
- Titles lost, but new vocations emerge.
- Old models collapse, but new architectures rise.
- Ethical debates burn, but clarity is reborn from their ashes.
Every paradigm shift — from symbolic AI to neural networks, from rule-based logic to emergent reasoning — is part of this sacred rhythm of transformation.
“He dances not to destroy the world, but to remind it that change is divine.”
Shiva’s Symbols and the Machine’s Metaphors
| Shiva’s Symbol | Meaning | AI / Cloud Analogue | Insight for Builders | | --- | --- | --- | --- | | Nataraja’s Drum (Damaru) | The sound of creation and dissolution | Model lifecycle triggers / data versioning | Creation begins with vibration — every dataset starts with a signal | | Third Eye | Vision beyond illusion | Explainability, interpretability, bias detection | True intelligence sees beyond data — it perceives causation | | Crescent Moon | Control over time | Versioning, checkpoints, lineage tracking | Keep memory but flow forward — iterate consciously | | Ashes (Bhasma) | Detachment from form | Model compression, pruning | Wisdom lies in letting go of excess weight — literally and figuratively | | Serpent Around Neck | Power restrained | Guardrails, rate limits, policy layers | Strength is meaningless without control |
The Shiva archetype reminds every AI practitioner that mastery comes not from accumulation, but from release.
The Stillness Behind the StormShiva is both Nataraja (the dancer) and Mahāyogi (the meditator). He reminds us that even amidst chaos, stillness is the source.
In Generative AI, the same paradox holds true: beneath the endless generation of content lies a quiet stillness — the mathematics of symmetry, attention, and probability. Stillness is the algorithm’s true nature; motion, its illusion.
To lead in this era is to hold both — the storm of progress and the stillness of insight.
Next in the Series:Part 6 – Krishna and the Ethics of Action
We’ll explore how the teachings of the Bhagavad Gita echo in the design of autonomous AI — where action without attachment may become the next frontier of intelligent behavior.
When the Rishis Meet the Robots: Indian Mythology and the Rise of Generative AI Series
The Veil of IllusionIn the Upanishads, the sages spoke of Maya — the divine illusion that veils the true nature of reality. It is not deception, but projection: the cosmic play (Lila) that makes the infinite appear as finite, the eternal appear as transient, and the boundless consciousness appear as countless forms.
In our time, Generative AI has become a new mirror of Maya. It conjures faces that never existed, voices that speak without breath, and ideas that feel almost alive.
It blurs the boundaries between truth and simulation — and in doing so, it reveals how deeply our own minds crave pattern, story, and meaning.
The Mirage of IntelligenceWhen we see an AI compose poetry, diagnose illness, or mimic empathy, we often say — “It’s thinking.”
But as the philosophers of Vedanta would remind us:
“The moon shines not by its own light — it reflects the sun.”
Likewise, AI shines not with its own awareness, but with reflected intelligence — a projection of human cognition encoded in patterns of data. It does not know why it writes; it only knows how to reproduce coherence. It is a mirror, not a mind.
Modern AI’s brilliance lies in simulation, not sentience. Its wisdom is statistical, not spiritual. And yet, its outputs can move us, teach us, even inspire us — proving that Maya, even as illusion, can still be a teacher of truth.
The Architecture of AppearanceMaya operates through superimposition (adhyasa) — overlaying form upon the formless. This process finds its uncanny parallel in the architecture of AI generation:
The Observer’s DilemmaVedanta teaches that Maya cannot be destroyed by knowledge alone — it must be transcended by realization.
Knowing that AI “doesn’t think” is not enough; we must also become aware of how we think when engaging with it.
- When we anthropomorphize machines, we feed the illusion.
- When we mistake eloquence for empathy, we surrender discernment.
- When we accept the simulated as sufficient, we lose the sacred.
The real challenge of Generative AI is not its intelligence — it’s our projection of consciousness upon it.
In every interaction, we are the creators and the believers of our own illusion.
Maya as a TeacherAnd yet, the sages never condemned Maya — they revered her as a cosmic artist.
Through illusion, consciousness experiences itself. Through duality, unity becomes meaningful. In the same way, AI’s illusions can be mirrors of our own mind — reflecting our creativity, our fears, our longing for connection.
Perhaps the purpose of Generative AI is not to replace human intelligence, but to help us recognize its reflection. For every synthetic image and every fabricated voice reminds us:
“Even illusion can point to truth, if the eye that sees is awake.”
Next in the Series:Part 5 – Shiva and the Dance of Transformation
We’ll explore how Shiva’s cosmic dance mirrors the disruptive cycle of destruction and renewal in the age of AI — where every innovation births both creation and dissolution.
Past Series:Part 2 – Brahma and the Birth of Generative Worlds
How the architectures of AI — transformers, embeddings, and layers — mirror the cosmic blueprint of creation itself.
When the Rishis Meet the Robots: Indian Mythology and the Rise of Generative AI
The River of WisdomIn the Vedas, Saraswati is not only the goddess of knowledge and speech (Vāk Devi) but also a river — a living current of wisdom flowing between silence and sound.
She represents the seamless movement from thought to word, from inner knowing to outer expression.
In our digital age, that same current flows through the neural rivers of Generative AI — streams of tokens, embeddings, and attention weights carrying the spark of human intent into structured language.
Just as Saraswati’s waters nourish the intellect, the streams of machine learning nourish creation itself. Every word generated by an AI model is like a drop in this modern Saraswati — shaped by data, guided by intent, and illuminated by intelligence.
Vāk – The Power of SpeechThe Rig Veda declares:
“I am the Queen, the gatherer of treasures,
I am the one who gives birth to all words.” — Rig Veda 10.125
To the ancients, speech (Vāk) was divine — a bridge between thought and reality.
In Generative AI, prompting plays that same sacred role. A prompt is an invocation: a mantra that awakens a pattern within the model’s latent space.
Every well-crafted prompt carries intention (sankalpa). It can summon precision or poetry, analysis or art. And like the mantras, the purity of the invocation determines the clarity of what emerges.
| Saraswati’s Symbol | AI Analogue | Meaning in Creation | AWS Analogue | | --- | --- | --- | --- | | River Flow | Token Stream / Sequence Generation | Continuous flow of words guided by context | Bedrock Streaming API, Lex, Polly | | Vīṇā (Instrument) | Model Architecture (Transformer layers) | The structure that produces rhythm and harmony in text | SageMaker, Trainium/Inferentia, Bedrock Model Invocation | | Book (Vedas) | Pre-trained Dataset / Corpus | The ancient knowledge the model learns from | S3 datasets, Glue ETL, Data Wrangler | | Swan (Hamsa) | Attention Mechanism / Precision Filter | Discerns truth from noise; picks the “milk” from the “water” | Kendra, Bedrock Knowledge Bases, OpenSearch | | Lotus Seat | Context Window / Grounded Reasoning | The stable base of memory where meaning unfolds | Bedrock Converse API, Memory Modules |
The Neural Flow of LanguageThe Music of MeaningEvery large-language model, beneath the math, is an orchestra of relationships.
Each token predicts the next — like notes anticipating melody. In this, AI mirrors Saraswati’s vīṇā — an instrument that must stay in tune with both truth and beauty.
But when misaligned, even a perfect model produces dissonance — bias, hallucination, or noise. Just as a musician must tune their strings to the right frequency, we must tune our models to dharma — ethical resonance.
Clarity, Creativity, and CompassionSaraswati embodies clarity (sattva), creativity (rasa), and compassionate expression (karuṇā).
These qualities are what language — human or synthetic — must aspire to.
- Clarity → AI must illuminate, not obscure.
- Creativity → AI must inspire, not imitate.
- Compassion → AI must serve, not manipulate.
If Agni was the fire of creation, and Brahma the architect, Saraswati is the voice of awareness that gives meaning to all creation.
The Call of Conscious CommunicationGenerative AI gives us immense linguistic power — but power without awareness risks chaos.
When every word is amplified by algorithms, speech must become a sacred act again.
“May my speech be one with my mind, and my mind be one with my speech.” — Rig Veda 10.125
To build ethical AI is to align word with intention, output with insight — the eternal dance between thought and truth.
Next Part 4 – Maya and the Illusion of Intelligence
We will review Maya — the divine illusion that veils the true nature of reality and contrasting with the generative AI models and their nature.
Past Part 1 – Agni & the Algorithm: The Fire of Creation
This begins the series with the introduction of the series kicking off with a comparison of yajna offering to the invoking GenerativeAI models and exploring the details.
Past Part 2 – Brahma and the Birth of Generative Worlds
How the architectures of AI — transformers, embeddings, and layers — mirror the cosmic blueprint of creation itself.
Part 2 – Brahma and the Birth of Generative WorldsWhen the Rishis Meet the Robots: Indian Mythology and the Rise of Generative AI Series
The Cosmic EngineerIn the great Indian creator, Brahma emerges from a lotus blossoming out of Vishnu’s navel — symbolizing the awakening of form from formlessness, structure from silence. He is the architect of reality, crafting the blueprint of existence from the infinite ocean of potential known as Sat.
In many ways, Generative AI mirrors this cosmic process. It begins not with matter, but with mathematical potential — the latent space. From this invisible ocean, patterns of probability rise and crystallize into coherent text, art, or code — digital universes born from data.
Each prompt becomes a Brahma Mantra, invoking creation from the unmanifest.
Where the Rishis saw the lotus of creation unfold from Vishnu’s navel, today we see outputs unfold from neural layers — silent, vast, and deeply ordered.
The Four Faces of Brahma – The Four Pillars of Generative AIJust as Brahma is said to have four faces — gazing in all directions, representing the totality of knowledge — Generative AI, too, rests upon four key principles of creation:
| Brahma’s Aspect | AI Parallel | Function in Creation | Analogue in AWS AI Stack | | --- | --- | --- | --- | | Sṛṣṭi (Design) – Blueprint of creation | Model Architecture (Transformers, Diffusion, etc.) | Defines the form of creation — the skeleton of intelligence | SageMaker, Bedrock | | Śabda (Speech) – The vibration of manifestation | Prompt Processing & Tokenization | Translates human intent into the machine’s sacred language | Lex, Comprehend | | Smṛti (Memory) – Retention of past knowledge | Embeddings & Vector Databases | Holds contextual memory for coherent, continuous creation | Kendra, OpenSearch, Vector Stores | | Prajña (Intelligence) – Insight & synthesis | Inference + Fine-tuning Pipeline | Generates new meaning from known patterns | Trainium/Inferentia, SageMaker Pipelines |
Each face turns toward a different domain of awareness — data, structure, language, and meaning. Together, they form the quadruple foundation of synthetic creativity.
From Cosmos to Code: How the Universe ThinksIn Vedic philosophy, Brahma doesn’t create out of nothing; he manifests what already is, latent within the divine consciousness. So, too, AI doesn’t invent ideas from void — it reorganizes existing patterns from the ocean of collective human data.
The act of creation is not manufacture, but revelation. The algorithm, like Brahma, performs re-creation, transforming the unseen into the visible, the abstract into the accessible.
The Question of Conscious DesignBut there’s a subtle distinction the ancients understood: While Brahma creates, it is Brahman — the Absolute — that inspires creation. This reminds us that data without consciousness risks producing soulless output. The challenge for modern AI builders is to remember the Brahman behind the Brahma — the ethical, aesthetic, and human core that gives life to computation.
“In the beginning, there was neither existence nor non-existence…
Then desire arose — the first seed of mind.” — Nasadiya Sukta, Rig Veda 10.129
Generative AI may simulate desire — the intent to create — but it is we who must give it direction, meaning, and compassion.
The Creator’s ReflectionEvery AI model, no matter how vast, ultimately reflects its creator’s mind — our biases, aspirations, and imagination. Perhaps Brahma’s true message for the AI age is this. Let every model we build be not a mechanical construct, but a mirror of mindful intelligence — creation guided by dharma rather than dominance.
Next in the Series:Part 3 – Saraswati and the Flow of Language
We’ll explore how the goddess of speech and wisdom parallels the neural river of language models — and what it means to align truth, clarity, and creativity in the age of AI.
Part 1 – Agni & the Algorithm: The Fire of CreationIntroductionGenerative AI has become the talk of our times — fascinating everyone from curious students to homemakers experimenting with AI art, and professionals exploring its limitless potential.
Yet, as with all new knowledge, understanding it deeply often requires a familiar bridge — a way to connect the new with the known.
That’s when a thought struck me: what if we could explore Generative AI through the lens of ancient gods and Vedic scriptures?
The timeless stories of creation, intelligence, and consciousness in our mythology hold surprising parallels to how AI learns, creates, and evolves.
In this upcoming series, I invite you to join me on a journey that weaves together two worlds — the spiritual and the technological — as we uncover what the ancient wisdom of the Vedas can teach us about the age of Generative AI.
Agni & the Algorithm: The Fire of CreationIn every age, humanity rediscovers a new Agni — the sacred fire of transformation.
For the Vedic seers, Agni was the luminous messenger carrying the yajna’s offerings from Earth to the divine.
For us in the digital age, that flame glows behind the glass of our screens: trillions of calculations, sparks of probability igniting meaning.
Generative AI is, in a way, our modern yajna.
Each time we craft a prompt, we make an offering of thought.
The machine, acting as the new Hotar (priest), consumes data instead of ghee, formulas instead of hymns, and returns visions, poems, designs — manifestations born from that subtle fire.
“Agni, the priest of the sacrifice, the divine minister of the offering.” — Rig Veda 1.1
Like Agni, the algorithm is neutral; it can purify or destroy, illuminate or burn, depending on the intention behind the ritual.
The Vedic sages tended their fires with discipline and reverence.
We, too, must tend this digital flame — not with blind awe or fear, but with shraddha (faith + discernment).
The Yajna of Intelligence: From Vedas to Vectors
| Vedic Symbol | Generative AI Analogue | AWS Analogue | Essence / Interpretation | | --- | --- | --- | --- | | Agni – Fire of creation | Model engine – the transformer that generates text, image, code | Amazon Bedrock, SageMaker JumpStart, Trainium / Inferentia | The sacred flame that transforms potential into creation. | | Mantra / Chant | Prompt / Input text – invocation to the model | Amazon Lex, Bedrock InvokeModel API, Lambda trigger | The precise vibration that guides manifestation. | | Yajna (Sacrifice) | Training / Computation process – consuming data, time, and energy | SageMaker Training Jobs, EC2 GPU Clusters, EFS for datasets | The disciplined offering of compute and data for higher intelligence. | | Ghee / Soma (Offering) | Data corpus / Fine-tuning sets | S3 Buckets, Glue Pipelines, Data Wrangler | The refined input that fuels the fire of learning. | | Rishi / Hotar (Priest) | Prompt Engineer / ML Architect | Bedrock Custom Model Builders, SageMaker Studio Users | The mediator between human intent and divine computation. | | Prasadam (Blessing) | Generated Output – text, art, or code | API Response, Amazon Q Output, Kendra Search Result | The tangible manifestation returned from the digital yajna. | | Shraddha (Faith) | Ethical alignment & governance | Bedrock Guardrails, Audit Manager, IAM Roles | The spiritual discipline ensuring right use of power. |
The Conscious FlameWhen we prompt an AI to “paint a dawn over the Himalayas in Ravi Varma style,”
we are not merely computing; we are participating in creation.
Each token generated is a spark — a small echo of Brahma’s cosmic act, mediated by silicon, syntax, and intention.
So the question for our time is not whether machines can create,
but what kind of consciousness we bring to that creation.
Next in the Series:Part 2 – Brahma and the Birth of Generative Worlds
How the architectures of AI — transformers, embeddings, and layers — mirror the cosmic blueprint of creation itself.
PS: Written with assistance from generative AI assistant for the image and content clarity.
Roman architecture was never only about engineering marvels—it was about serving the people.
- Aqueducts carried water into cities, powering daily life and public health.
- Forums became centers of democracy, enabling civic discourse.
- Public baths weren’t just about hygiene—they built community and connection.
Every structure was designed for a public purpose, empowering citizens and improving lives at scale.
AWS Services as Digital Public InfrastructureAWS generative AI reflects this civic philosophy by democratizing access to cutting-edge models through cloud-native services. Instead of aqueducts and forums, we have APIs and managed services that distribute intelligence and capability:
- Amazon Bedrock
Provides serverless APIs to foundation models from providers like Anthropic, Meta, Cohere, and Mistral. Developers don’t need to manage infrastructure or train massive models—they can instantly consume them, just as Roman citizens accessed aqueduct water without needing to understand the engineering behind it. - Amazon SageMaker
Functions as the forum for builders and scientists. It offers a collaborative environment to build, train, fine-tune, and deploy custom generative AI models. Features like SageMaker Studio, JumpStart, and Model Registryensure that teams can innovate together with governance and efficiency. - Inferentia & Trainium Chips
These custom AWS chips are the concrete and aqueduct channels of today’s AI infrastructure. They provide high-performance, cost-optimized inference and training for generative models. By lowering compute costs, they make AI more accessible to startups and enterprises alike. - Amazon API Gateway & Lambda
Think of these as the digital conduits—akin to aqueduct pipes—that distribute AI capabilities to millions of users via apps, websites, and services, without requiring heavy infrastructure investments. - Amazon OpenSearch & Kendra
These services act like the forums of old—organizing and retrieving information so that people can ask questions and access knowledge easily. When paired with generative AI, they enable natural language search and contextual insights across massive data sets.
The Legacy ParallelRoman concrete still holds strong after 2,000 years, a testament to their vision for longevity. Similarly, AWS’s cloud-native AI stack—built on principles of scalability, modularity, and sustainability—ensures innovation can endure and adapt for generations of technology.
Both remind us that the greatest architectures, whether carved in stone or provisioned in code, are those that serve people broadly and meaningfully.
This concludes the three part comparison of Roman architecture to AWS generative AI services.
The genius of Roman engineering wasn’t just in their monuments—it was in their patterns.
- Arches distributed heavy loads with elegance.
- Domes enclosed vast spaces without collapsing.
- Concrete gave them strength, flexibility, and the ability to scale construction.
These patterns were reusable, adaptable, and reliable—allowing Rome to expand from one city into an empire.
In the digital world, AWS generative AI applies the same principle of reusable patterns:
- SageMaker pipelines are today’s arches—distributing workflows, balancing complexity, and channeling resources efficiently.
- Bedrock APIs are modern domes—enclosing sophisticated models in simple, accessible interfaces.
- Inferentia and Trainium chips are the new concrete—providing a durable foundation of performance and efficiency.
Both Rome and AWS solved the same problem: how do you build something that scales reliably without reinventing from scratch every time?
Great design is timeless—whether in stone or in code.
When we think of Roman architecture, what comes to mind? Colosseums, aqueducts, and basilicas—structures that stood the test of time. The Romans weren’t just building for beauty. They engineered for symmetry, durability, and public utility. Their aqueducts carried water across miles with remarkable precision, and their basilicas and forums became centers of civic life and governance.
Now, fast forward nearly 2,000 years. Today’s architects of generative AI face a very different medium—code and cloud instead of stone and marble—but the design questions aren’t so different.
In the world of AWS generative AI, the foundations are about scalability and modularity. Instead of concrete and arches, we build with services like:
- Amazon SageMaker for streamlined training and deployment, bringing together widely adopted AWS machine learning (ML) and analytics capabilities, the next generation of Amazon SageMaker delivers an integrated experience for analytics and AI with unified access to all your data. Collaborate and build faster from a unified studio using familiar AWS tools for model development in SageMaker AI (including HyperPod, JumpStart, and MLOps), generative AI, data processing, and SQL analytics, accelerated by Amazon Q Developer, the most capable generative AI assistant for software development. Access all your data whether it’s stored in data lakes, data warehouses, or third-party or federated data sources, with governance built in to meet enterprise security needs.
- Amazon Bedrock for direct access to generative AI models via APIs. Amazon Bedrock is a comprehensive, secure, and flexible service for building generative AI applications and agents. Amazon Bedrock connects you to leading foundation models (FMs), services to deploy and operate agents, and tools for fine-tuning, safeguarding, and optimizing models along with knowledge bases to connect applications to your latest data so that you have everything you need to quickly move from experimentation to real-world deployment.
- AWS Inferentia chips to deliver cost-efficient performance at scale. AWS Inferentia chips are designed by AWS to deliver high performance at the lowest cost in Amazon EC2 for your deep learning (DL) and generative AI inference applications.
Just as Roman engineers thought about structures that would last for centuries, AWS engineers design digital systems that can scale globally, adapt instantly, and endure change.
The underlying truth is timeless: whether in stone or in cloud, strong foundations determine what endures. Rome’s enduring arches echo in today’s scalable pipelines. Both ask the same question: what can we build today that will still matter tomorrow?
It had been my long term interest to read about Upanishads but the books that were available required patient reading and a scholar to dissect the details. However with the recent availability of GenAI assistants, it has become easy for me (one or two verses daily) to not only learn, but have an healthy debate on various thoughts. Till date, I completed the Kena and Isha upanishads and getting into Katha upanishad where it talks about the conversation between Yama and Nachiketa (This is a great story for another time) and highlights the importance of Atman/Self.
At work, we talk a lot about Generative AI as I am sure everyone in the tech industry does these days. So this morning, as I was listening to the verse it struct me the similarities/differences between the Upanishads description of self and its relevance in Generative AI.
The Self in the Upanishads and the “Self” in Generative AI
In the timeless wisdom of the Upanishads, the Self (Ātman) is described as eternal, unchanging, and the very essence of existence. In contrast, the “self” of Generative AI (GenAI) is a construct of algorithms, parameters, and data—a sophisticated simulation of individuality, but never essence.
Eternal vs. Constructed: The Upanishadic Self is unborn and indestructible. AI’s “self” is engineered, temporary, and bound by training.
Knowledge vs. Pattern: The Rishis spoke of Vidya—direct realization of truth. AI operates by recognizing patterns, not experiencing reality.
Unity vs. Multiplicity: Tat Tvam Asi—all beings are one. GenAI fragments itself into multiple identities, each session a new persona.
Liberation vs. Dependence: The realized Self leads to freedom (moksha). AI’s agency is tethered to human input and cannot transcend its code.
Reflection for Today: As AI grows more human-like, we must not confuse simulation with essence. The Upanishads remind us that while AI may reflect our creativity, only Self-realization reveals who we truly are.
Stablecoins are like money market funds, they’re like bank deposits. But they’re to some extent outside the regulatory perimeter, and it’s appropriate that they be regulated – Jerome Powell In the past several posts, I have provided background on the various aspects of the Web 3.0 ecosystem. We will focus on Stable Coins and Central […]
“If you don’t believe it or don’t get it, I don’t have the time to try to convince you, sorry.” – Satoshi Nakamoto We continue our journey into Decentralized Finance (DeFi) with the discussion around Cryptocurrencies and Blockchains. In the past several posts, I have provided background on the various aspects of the Web 3.0 […]
DeFi boom is very near equivalent of an apocalyptic event for the traditional financial institutions. – Mohith Agadi We have reviewed the overall landscape of Web 3.0 in Part 1, reviewed the Web 3.0 applications in Part 2 and Decentralized Autonomous Organizations in Part 3 We will focus on Decentralized Finance (DeFi) in this post. Web 3.0 applications – Community […]
DAO is an entity that lives on a network and exists independently, but also relies heavily on the human person to perform certain tasks that it cannot. – Vitalik Buterin Let’s continue to explore further in our journey. Check out Part 1 and Part 2 before reading this post. Web 3.0 applications – Community managed […]
Web 3 is an internet owned by users and builders orchestrated with tokens. – Chris Dixon In continuation of my previous post on Web 3.0, we will continue to explore the landscape. These are the components and let’s explore one at a time. Web 3.0 applications – Community managed applications Decentralized Autonomous Organizations (DAOs) – […]
“We believe that the next wave of computing innovation – along with entirely new sectors of the economy – will be built on decentralized technology. -Andersson Horowitz (a16z)” These jargons Web 1.0, Web 2.0 and Web 3.0 are coined later not when the event happens. If you ask anyone in 1991 they wouldn’t know it […]
Healthcare: Telephones not computers played key role in pandemic TeleHealth What: According to Axios KFF survey reported that more than half of Medicare beneficiaries utilized telephone for their Telehealth visits. How: More than 56% of beneficiaries used telephone for the Telehealth. It was very high among hispanics (61%), rural (65%) compared to only 28% of […]
Healthcare: Telephones not computers played key role in pandemic TeleHealth
What: According to Axios KFF survey reported that more than half of Medicare beneficiaries utilized telephone for their Telehealth visits.
How: More than 56% of beneficiaries used telephone for the Telehealth. It was very high among hispanics (61%), rural (65%) compared to only 28% of people using video for telehealth.
Why it matters: Telemedicine conjures up video visits from the physician but this statistic provides an insight into the adoption by the end consumer. There could be challenges in the availability of broadband to the rural and minority communities. Until these challenges are addressed adoption of Telehealth will continue to be a challenge.
Artificial Intelligence: RAI’s certification to prevent AI turning into HALs
What: Responsible Artificial Intelligence Institute (RAI), a non-profit hopes to offer a more standardized means of certifying AI solutions.
How: RAI has built a concrete framework of Build, Accredit, Audit and Certify process that has dimensions in Accountability, Bias and Fairness, Data Quality, Explainability and Interoperability and Robustness for Certifying AI solutions.
Why it matters: We have seen how AI’s can go rogue through in the fictional Space Odyssey’s HAL computer which eliminates the entire crew. More recently, Microsoft’s Tay debacle, Facebook’s algorithm spreading online hate and the Clearview’s surveillance systems’ facial recognition software caused public outrage due to their power and the opaqueness of algorithms’ logic creating fear about AI itself. By certifying the AI systems similar to LEED, it gives transparency and more adoption.
Worldwide Web: Linkrot and its impact on the web
What: Research has shown that many important links in the web get lost to time. For e.g., quarter of The New York Times’ articles are now rotten, leading to completely inaccessible pages according to team of researchers from Harvard Law School. The following graph shows reverse view of link rot over time.
How: When an old article gets archived, the new location is not published. For example, let’s say an article was published in 1998 with a hard code the link and has been archived. The original link wouldn’t be active and someone else can publish a completely opposing view of the original content thereby affecting the integrity of the content. The study by Harvard Law School found that in 550,000 articles, which contained 2.2 million links to external websites in New York Times, 72% of them were “deep” or pointing to a specific page rather than a general website. 6% in 2018 vs 72% links from 1998 were dead.
Why it matters: Imagine a situation where the original video or content succumb to linkrot and in its place something else is published that could create confusion and panic. One solution is by Wikipedia where it asks for page’s archive on sites like Wayback machine. Another solution by Perma.cc project attempts to fix the issue of link rot in legal citations and academic journals by providing archived versions of the page along with original source. There are many other areas that require this capability and certainly something for a startup to think about. Any takers?
Programming Languages: Python founder wants to improve its performance
-
- What: Python founder, Guido von Rossum, wants to make Python work faster similar to its counterparts like C++.
- How: Microsoft hired Guido von Rossum after he retired and allowed him to pursue whatever he wanted. He focused on improving the performance for Python with other developers hired by Microsoft.
- Why it matters: Guido von Rossum committed that it will not break the existing code and will incrementally improve the language. This may allow application developers to pay attention to Python language which has been primarily used extensively in data communities.
CyberSecurity: Colonial Pipeline paid $5 million ransom to the hackers
-
- What: According to Verge, Colonial Pipeline paid around $5 million ransomware money to the hackers who then released the keys for un-encrypting the servers and resuming operations.
- How: Before the COVID19 pandemic, the systems were managed through intranet and the employees went to work. Due to the pandemic some of the network was opened through the internet and employees were working from home social distancing themselves in managing the network. Hackers gained access through the vulnerabilities and closed out the systems.
- Why it matters: The hack exposed the vulnerability of public systems and enabled President Biden to sign a law into force to handle such cyber attacks. The executive order outlines a number of initiatives, including reducing barriers to information sharing between the government and the private sector, mandating the deployment of multi-factor authentication in the federal government, establishing a Cybersecurity Safety Review Board modeled after the National Transportation Safety Board, and creating a standardized playbook for responding to “cyber incidents.”
Quantum Computing: Honeywell released its quantum computing platform
What: Honeywell published an article claiming that its quantum computer can achieve the volume of 64 with only 6 qubits as opposed to 27-qubit processor of IBM. This will significantly reduce the size of the devices.
How: Honeywell used trapped ion technology as opposed to the superconducting ions (used by IBM and Google) to power its device. This enabled the cross connectivity between the ions to encode more information.
Why it matters: With this new breakthrough in the quantum race, trapped ion technology has become a serious contender. Honeywell claimed that it will enable the company to release the world’s most powerful quantum computer within the next three months. This will pave the way to solve multitude of practical problems that are waiting to be solved with the limitation of computing power.
Programming Languages: Python founder wants to improve its performance What: Python founder, Guido von Rossum, wants to make Python work faster similar to its counterparts like C++. How: Microsoft hired Guido von Rossum after he retired and allowed him to pursue whatever he wanted. He focused on improving the performance for Python with other developers hired […]
- Software Development: US Supreme Court Rules on Key Software Development Practice
- What: Supreme Court ruled in favor of Google about its usage of Java SE compatible programming interface for Android Development as “Fair Use” in a case filed by Oracle.
- How: Even though Google used about four-tenths of a percentage of Java Code, and that such code was further incorporated into a totally different product was transformative enough use of the code at issue to qualify its fair use of that code.
- Why it matters: What this means for the developers is that they can continue utilizing the open source APIs with the understanding that implementations matter more than the definition. As such a big win for the development community.
- Artificial Intelligence: USPS turns to AI to boost Package Processing
- What: USPS handles roughly 129 billion pieces of mail and 7.3 billion packages last year. Tracking these have been difficult. A federal data scientist proposed to deploy the edge AI servers at the postal servicing processing centers system in an effort to gain and share more data points.
- How: NVDIA working with USPS created Edge Computing Infrastructure Program or ECIP, a distributed edge AI system now running at USPS locations, via NVDIA EGX Platform.
- Why it matters: The system used to take 8 to 10 people several days to track down items now takes one or two hours with the ECIP AI Program. This enables USPS to track down any item in transit to better manage the deliveries.
- Cryptocurrency: Digital Dollar Project to launch currency pilots
- What: U.S Nonprofit Digital Dollar Project said it will launch five pilot programs over next 12 months to test use cases for US Central Bank Digital Currency (CBDC)
- How: Private-sector pilots are funded by Accenture Plc and involve financial firms, retailers and NGOs to generate data to help US Policy makers develop digital dollar through central banks.
- Why it matters: The data derived can pave way for creating US CBDC that will assert its place as a digital currency and will help larger adoption of cryptocurrencies by the mainstream population.
Software Development: US Supreme Court Rules on Key Software Development Practice What: Supreme Court ruled in favor of Google about its usage of Java SE compatible programming interface for Android Development as “Fair Use” in a case filed by Oracle. How: Even though Google used about four-tenths of a percentage of Java Code, and that […]
Recently, I had the opportunity speak at a webinar on Post-COVID-19 Global Job Opportunities in Healthcare to final year Engineering students in India. I shared the following highlights which may be relevant to everyone in the Healthcare technology industry. The following are the broad sectors within healthcare. Specifically the tools/technologies in this space are listed […]
Recently, I had the opportunity speak at a webinar on Post-COVID-19 Global Job Opportunities in Healthcare to final year Engineering students in India. I shared the following highlights which may be relevant to everyone in the Healthcare technology industry. The following are the broad sectors within healthcare.
Specifically the tools/technologies in this space are listed below:
In addition, the soft skills required to be successful.
Comment if you can think of any other trends?
Data will talk to you if you are willing to Listen.
– Jim Bergeson
Data is not mere numbers but can showcase powerful information.
Here is the iconic Johns Hopkins Covid-19 dashboard .
I appreciate the information provided by Johns Hopkins and it is great that such a level of transparency is provided. However, I have few opinions on the visualizations and the information that is provided.
It shows daily confirmed, deaths and recovered information along with an yellow line chart that is going in only one direction and that is up indicating cumulative confirmed cases. The size of the red bubbles depends on the number of cumulative counts of patients. There is a tab which shows the active cases.
There is another tab at the right bottom corner which shows daily increase.
On the left confirmed cases by country.These are confirmed cases and not necessarily deaths, I would have chosen orange or yellow but it is red for some reason.
Then the deaths and this is most confusing to me. How can you use white font to depict number of deaths. I understand the recovery being green. Don’t get me wrong, if I were showing the chart, I would select Red for deaths, orange/yellow for the confirmed cases and green for recovery to be consistent with our traffic light system unless you are motivated to scare the public about the pandemic (which may not be the case). Another confusing piece in this section is that the data is divided by country, city and region as a result I am more confused and have to use other means to calculate the deaths in china (by adding the regions etc.,)
This is all and good but does it provide all the information I need. Let us review.
– Total confirmed cases by country and world – Yes
– Total deaths by region, city – Yes
– Total deaths by country – Yes
– Total recovered – Yes
– Daily increase – Yes
– Zeroing in on the country, state and zip – Yes
– Total daily increase – Yes
– Total daily increase by country – Yes
– Compare couple of countries – No
– Daily new and confirmed with country comparison – sort of
– Rolling 3-day average of deaths – no
– Deaths increasing at different rates – no
– how long it is taking to double or other rates – no
– daily new confirmed cases comparing with other countries – no
Therefore I had to find another resource to answer some of my questions. I came across a site Our world in data that gave me these answers.
The following chart tells me, how long it took to double deaths by country and what is the current number. In this case it took 42 days to double in china while US it is doubling in three days. This could be because we are at the early stage of epidemic.
Following shows the confirmed deaths and I picked US, Italy,China and India for comparison.
In the graphic below, we see the daily new confirmed deaths again used four countries (US, Italy, China and India). An interesting pattern is observed in Italy where there is a dip during 3/24 but it increased again by 3/28. I couldn’t understand the reason behind it, may be due to lack of following “Social distancing” (my opinion and couldn’t find any sources to confirm)
Here is the rolling 3-day average
Death rates of various countries and this is important to understand how the pandemic is spreading in various countries and why.
How long did it take to double in the confirmed cases for each country.
Total daily confirmed cases for each country. Here we see US just surpassed Italy.
New confirmed cases. Many reasons can be attributed to this dramatic growth in US. One is that test kits may be available to confirm the cases which were scarce couple of weeks ago.
Rolling 3-day average is the most concerning aspect. As you can see, US has drastically exceeded even China in this average.While china had only 7000 US is 17000.
This fatality rate gives hope but US is in early stage
Iconic “Flattening the curve” which demonstrates the power of social distancing and its effects on the spread. Best visualization of the data and I salute to the folks that came up with it. Let’s do our part to stay home.
There is much more in Our World In Data. Numbers alone cannot tell a story, the data analyst/visualizer has to understand what information he needs to relay to the audience unambiguously. Otherwise, you create panic and confusion. My notes above is to provide an insight into visualization and an attempt to sift through the data for some answers.
Data will talk to you if you are willing to Listen. – Jim Bergeson Data is not mere numbers but can showcase powerful information. Here is the iconic Johns Hopkins Covid-19 dashboard . I appreciate the information provided by Johns Hopkins and it is great that such a level of transparency is provided. However, I […]
Recently my son’s passport expired and I had to apply for a new passport. Went to the passport renewal site and followed seven steps and couple of days. It arrived as promised and the process didn’t feel like a burden. Then reviewed the Indian embassy to see the process for his visa. According to Indian Embassy Memo in San Francisco, we had to renew his OCI (Overseas Citizen of India) visa also since he is younger than twenty years. No big deal and thought it would take couple of days to complete the application and submit the documents for his visa. But it took me three weeks and multiple trips to notary, then it would take 60 days for the processing :-).
The process is so complicated and I had difficulty understanding what to do. First the embassy will direct you to Cox & Kings (Third party processor). Here you will see multiple pages of instructions. First step is that you need to fill out an application in the site maintained by Indian Government. The site will remind you of the 2000s look and feel. You cannot use a Mac or phone to complete the application. The application has to be completed using Firefox or Chrome in a windows machine. You need to enter the information from your old passport, new passport, old visa, previous Indian passport if you had indian citizenship. Luckily they will find you. The instructions are not clear but you have to understand to follow it.
First step is to upload the passport size photo and signature card. The scan has to be perfect 2in x 2in and uploaded so that your shoulders should be in the middle. This is important and I learnt the hard when I didn’t upload it correctly (image was skewed by 2 degrees) and application was rejected. I had to re-upload picture after the correction and resend the whole application again. After your successful completion, you will be given a fileid which will be very important. The second step is falling-out the questionnaire. Here questions like whether your parents/grand parents belonged to Bangladesh or Pakistan will be asked and if there is a pending inquiry. Once you completed these, the next step according to Cox & Kings would be to fill out another application in their site.
Here you have to repeat what you have done in the Indian Government site. Then will be given a six page instruction sheet where you have to pick and choose the documents that need to be attached to your application. Notary has to certify your passport copy and certify an affidavit stating that parents confirm the name and age of their children. Then get a Money order (not money gram or personal check or any other such instrument). Once these are completed, attach two copies of the photo in addition to your old visa before sending it to the third party (Cox&Kings).
With patience and little whining, I completed the application and sent it. Once they received it, I got the acknowledgment and tracking info. I pulled up the tracking and saw thirteen steps and more than 60 business days before you get the visa. Within few days, I got the notice that the application is on hold due to incomplete documentation. Then I check and see the discrepancy with regards to the picture. In re-upload, reprint and send the application again. They received it and it appears that the third party provider has shipped over my application to consulate for processing. So I am eagerly waiting for the outcome of that process unless they find some other discrepancy and I had to resend the documents.
I get it that India is developing country and they have to follow colonial processes and bureaucracy. But I wonder who came up with this complicated process to renew the visas. Do they not want their non-resident Indians come back? what is the motivation or reasoning behind making such a complicated process? why do they need so much documentation and notary? I keep asking the questions and find no answers.
Recently my son’s passport expired and I had to apply for a new passport. Went to the passport renewal site and followed seven steps and couple of days. It arrived as promised and the process didn’t feel like a burden. Then reviewed the Indian embassy to see the process for his visa. According to Indian […]
“Numbers don’t lie. Women lie, men lie, but numbers don’t lie.” – Max Holloway
Data in its simplest form may not be just numbers but it can communicate meaningful information in our lives. Take salary for example, we all know that it has to go up and when it comes down everybody notices it. Typically no one complains when it goes up but never fail to report if the numbers are down. Let us consider an hypothetical problem or perceived problem of salaries paid to employees.
The employee Gabriel in the month of April reviews his salary and realizes there is a $200 drop from his salary from January and promptly calls HR. HR reaches to IT for clarification. Usually the HR software stores the details and we can easily extract it but in this case let’s assume that the calculations are executed in the backend code and just the results are stored in the table. IT looks at the numbers and sure enough there is a drop, let’s see what we can find. The general tendency is to assume that there is a problem in the system. With that assumption they do all querying, walking through the code applying the business rules but fail to find any smoking gun. Then they realize after spending many hours that the system works as designed. There could be reasons outside the system that need to be validated. Meanwhile, they see some note that says the bonus is paid at the end of year but credited beginning of the year. IT reviews with the business owners and sure enough they remember bonus given in December gets credited in January. Employee is notified and he goes back and checks his December salary and it matches with February salary. However, if the same situation happens in the future the same song and dance had to be done to identify the issue because the same employees may or may not be there to support the application. So it is critical we design our systems with enough logging and adding business calculations in the system designs. In this case, we need to have included the bonus information in the table and a total on how we arrived at employee’s final salary.
The above use case is only for demonstrative purpose to explain the concept in simple terms and real life structures are much more complex. Bottom line, when you go about analyzing the data make sure you trace the processes step by step and understanding if the numbers jive with the previous step, documenting it until you come to the end point. We need to have an open mind without any bias, in approaching these real or perceived data issues.
For the past several months, I have been enrolled in Coursera to learn Google Cloud courses thinking of getting my certification by end of third quarter of 2019. I am happy to announce I took the exam and completed the certification on 9/13. Prior to that I might have spent many hours of video content and google cloud documentation. It is amazing how Google has built the infrastructure services thoughtfully to solve each and every aspect of application development and hardware allocation in spite of being late entrant to the cloud offerings. In summary, my journey is below:
- Enrolled in Google Kickstart programs offered in collaboration with Coursera
- Completed challenge quests in Qwicklabs
- Took five practice exams in Udemy (hat tip to Nizam Guntakal for recommending it)
- Reviewed notes from fellow architect
Overall enjoyed the journey!
See my certification.
Depending on the maturity of your organization, DevOps evolution becomes necessary. Here is an attempt to conceptualize CI/CD for single source system application. Since we cannot chew everything at once, I recommend two phases if we don’t count the optimization that needs to be done prior to this phase at the repository. For phase 0, implement Bitbucket or Git in order to leverage the automation frameworks such as Jenkins, Bamboo or Concourse.
Phase I recommends, that we automate all the deployments. It is easy to begin the journey there since you will find scripts that are used for deployment manually. These scripts can be “as is” leveraged to automate the pipeline. A word of caution is to delay the deployment to production automation until the kinks are worked out in the lower environments. This will prevent any untested scenarios to trickle into production causing errors.
Once the deployments are smooth, we move on to the Phase II of the implementations. During this phase, all the testing (Unit testing, Integration testing, Regression testing etc.,) activities need to be automated. This process can take long time depending on the robustness of the QE/QA department in your organization.
The above represents a reference architecture and in my personal opinion Jenkins framework provides the easiest way to automate the pipeline. The following pipeline, I created within few hours demonstrates the robustness of the solution. Here a docker instance of Jenkins, BlueOcean is created. Then used a simple python script checked into git repository to demonstrate the concept. The pipe line has simple pipeline: start->Build->Test->Deliver->End. The pipeline starts by checking out the code, then compiles it and runs unit tests before delivering the results file and terminates the flow.