An Ode to San Francisco
I have moved back to San Francisco after a hiatus of decades, returning to a city that I thought I knew, only to find that my memory was a map of a place that no longer exists.
Walking these streets again feels like visiting a house you grew up in, only to discover someone has rearranged the furniture and repainted the walls in colors you can’t quite name. I am walking through the ghosts of my early years here—past the corners where I used to buy coffee and the apartments where I used to dream—but the backdrop has shifted.
It is a strange and beautiful disorientation.
The world loves to measure San Francisco. It tries to capture this peninsula in spreadsheets and headlines. They measure it by its market caps, its office vacancy rates, its unicorns, and its crime statistics. They look at the city and see a balance sheet of booms and busts, trying to quantify the magic or the misery.
But from where I’m standing—winded by the hills and squinting through the fog—those measurements miss the point entirely. They are measuring the container, not the contents.
If you look strictly at the hardware, San Francisco is confusing. The buildings are rarely the most beautiful. The pavement is often cracked, jutting up at odd angles where tree roots have won the war against concrete. The infrastructure creaks; the paint is peeling on the Victorians; the brutalist towers are gray and imposing. If you were judging the city by its physical shell—by the smoothness of its roads or the polish of its facades—you would assume it was failing.
But San Francisco has never cared about the hardware. This is a city entirely devoted to the software.
The "vibe" here is a tangible thing, a humming voltage that exists independent of the architecture. It is the only place I know where the air itself feels kinetic. While other cities obsess over their skylines, San Francisco obsesses over its operating system—the invisible network of ambition, delusion, brilliance, and relentless optimism that runs underneath the streets.
It is the density of the "shared brains" that gets you. You can walk into a mediocre coffee shop on a cracked sidewalk and stumble into a conversation that is rewriting the future of biology or computing. There is a total lack of hierarchy and a surplus of curiosity. The quality of the people here isn't just about intelligence; it's about a specific kind of openness. It’s a collective agreement to suspend disbelief, to act as if the impossible is just a debugging issue away from working.
To be back here is to realize that I am not a veteran returning to his old stomping grounds. I am just another newcomer in a city of newcomers. I feel less like a resident and more like a student again, dazzled by the raw processing power of the people around me.
I have come to see that my own life has been a "caravan of selves"—a constant process of shedding who I was to become who I needed to be. And I think I love this city because it is doing the exact same thing. It doesn't care about preserving the pavement; it cares about upgrading the soul.
The Voice in the Machine
If you walk through the current landscape of AI, you have to cover your ears. It feels less like a technological revolution and more like a bazaar where a thousand merchants are shouting the exact same sales pitch. Everyone is selling speed. Everyone is selling text. Everyone is selling a faster way to generate a generic email.
But standing here, in the middle of this cacophony, I have decided to stop typing and start listening.
At Sentium AI, we are trying to do something that feels dangerously different. We aren’t interested in building another text-generator that sits passively in a browser tab. We are building conversational-first companions for business.
The distinction sounds subtle, but when you actually experience it, it feels like a fracture in reality. We are moving away from "chatbots" and toward Synthetic Business Consciousness.
Most AI today is just a very fancy autocomplete; it guesses the next word. But at Sentium, we are trying to engineer a mind that understands context, holds a memory, and—most crucially—can actually do things. These companions are voice-first. They don't just wait for a prompt; they listen. They understand the nuance of a pause; they get the intent behind the instruction.
And they don't just talk back. They execute.
It is a strange sensation to speak to a machine and feel it actually grasp the business logic, not as a database query, but as a concept. It feels like we are graduating from building tools to raising digital colleagues.
I admit, I am still the student here. I am watching this technology evolve with a mix of terror and delight. We are trying to bridge the gap between a static piece of software and a fluid, thinking companion. We are teaching the machine to hear the signal in the noise, to understand the order, and then—without needing to be micromanaged—to go out and make it happen.
We are building Sentium not to add to the noise, but to finally offer a voice that understands.
9-9-6 versus 1- 2- 3
Building Teams That Compound
The business world is currently enthralled in a dangerous love affair with "hustle culture." We eulogize the grind. We celebrate the sleepless nights. We scroll through LinkedIn posts glorifying leaders who sleep under their desks.
Nowhere is this more codified than in the infamous "996" work culture—working from 9:00 a.m. to 9:00 p.m., six days a week. It is touted by tech giants and ambitious startups alike as the necessary price of admission for success in the modern economy.
I have walked the long road of entrepreneurship. My journey began back in 1995 with Candid Marketing, moving through hospitality, real-world technologies, and now into the frontier of AI in 2024. I’ve carried the backpack along the winding path, hitting milestones and weathering storms.
Over these three decades of building companies and managing diverse teams, I have tried various models of productivity. I have seen the allure of brute-force labor.
And I am here to tell you that the world is wrong about 996.
It is a math equation that leads to diminishing returns, employee burnout, and a culture of "furious activity" rather than actual achievement.
As I’ve matured as a leader, I discovered a different path. It’s less about squeezing every drop of blood from a stone, and more about strategic leverage. I call it the 1 2 3 Formula. It is counterintuitive, it requires courage, but it is vastly superior to the exhausting grind of 996.
Here is why the future belongs to those who understand 1 2 3, and why 996 is a relic of the past.
The False Economy of 996
Exhaustion as a Metric
The premise of 996 is seductive in its simplicity: Input equals output. If you want more output, you simply demand more hours of input. It’s an industrial-age mindset applied to information-age work.
But knowledge work isn't an assembly line. You cannot simply crank up the speed of the conveyor belt to get more brilliant marketing campaigns or elegant code.
When you force a team into a 996 structure, you aren't getting 72 hours of peak productivity. You are getting perhaps 40 hours of actual work, diluted by 32 hours of exhaustion, resentment, mistakes, and "presenteeism"—the act of being at your desk just to show you are there.
I’ve seen the inside of 996 offices. They look like the "before" picture in a productivity makeover. The teams are disheveled. The energy is frantic, not focused. They are furiously working away on laptops, wires crossed, coffee cups overflowing, eyes glazed over. They are "unkempt" not just in appearance, but in spirit.
This culture bleeds talent. Your best people—the ones who know their worth—will leave. You are left with mercenaries who are there for the paycheck, or those trapped with few other options. 996 isn't a strategy for growth; it's a strategy for attrition.
The Epiphany
Leverage Over Labor
Early in my career, juggling clients like Coca-Cola and Microsoft, I felt the pressure of "too many clients, too little time." My instinct was to work harder, longer.
But as I transitioned from running a service company to building technology and AI platforms, I realized that linear effort doesn't scale. To hit the massive milestones, you need exponential output. You don't need more hours; you need better hours. You need leverage.
Leverage doesn't come from exhausting people. It comes from empowering them. This led to the development of the 1 2 3 Formula.
The 1 2 3 Formula
The Blueprint for Exponential Output
The 1 2 3 Formula challenges the traditional hiring mindset. It demands that leaders stop looking for "cheap labor" to fill seats and start investing in high-leverage talent.
Here is how it breaks down:
1. Hire ONE (The A-Player)
Forget hiring three average people to do a big job. Find the one person who is an absolute master of their craft. This is the "10x engineer," the visionary marketer, the operational wizard. They are hard to find, and they demand autonomy, but they are the foundation of this formula. You are hiring for capability, not just capacity.
2. Pay, Guide, and Love for TWO (The Investment)
This is where most companies fail. When you find that A-Player, you cannot treat them like an average employee.
-
Pay for Two: You must compensate them significantly above market rate. You might pay them double what an average person in that role makes. It sounds expensive until you look at the results.
-
Equip for Two: You don't just give them money; you give them the best tools, the best software, and the best training.
-
Love and Guide for Two: This is the human element. You provide intense mentorship, clear guidance, and genuine care for their well-being. You protect their time. You ensure they aren't bogged down in low-value tasks. You are investing heavily in their success.
You are essentially treating this one person as an entire high-functioning department.
3. Deliver for THREE (The Return)
This is the magic. A truly extraordinary talent, when highly compensated, fully equipped, deeply respected, and untethered from bureaucracy, will not just do the work of two people. They will deliver the output, creativity, and strategic value of three average workers.
Why? Because they don't make the mistakes that slow down average teams. They see shortcuts others miss. They automate tasks. They work with a state of flow and mastery that 996 workers can only dream of. Their work is clean, efficient, and impactful.
The Choice
Furious Activity vs. Strategic Mastery
The difference between these two models is stark.
The 996 Model is a war of attrition. It’s a crowded, sweaty room filled with tired people pushing boulders uphill. The boss feels good because everyone is "working hard," but the actual progress is incremental and agonizing.
The 1 2 3 Model is a SEAL team operation. It’s a clean, modern workspace characterized by mastery and efficiency. The team is smaller, better paid, well-rested, and highly focused. The atmosphere isn't frantic; it's intense but calm. They aren't just working; they are delivering exponential value.
As leaders, we have to stop eulogizing the grind. We need to stop confusing presence with productivity. The goal of business isn't to see who can stay at the office the longest. The goal is impact.
If you want linear growth and high turnover, choose 996.
If you want exponential growth, loyalty, and mastery, have the courage to choose the 1 2 3 Formula. Hire the best, invest deeply in them, and watch them outperform an army of exhausted souls.
My Journey and Insights
Welcome to my "Notes from the Road"—a curated collection of thoughts, reflections, and insights from my journey as a 4x founder and the visionary behind Sentium AI. These notes, drawn from my public posts and professional experiences, offer a glimpse into my thinking as I operate at the forefront of artificial intelligence, business, and human-computer interaction. I've organized them into thematic categories that explore the core principles driving my work and my vision for the future.
The Silent Crisis of Tribal Knowledge
Every organization believes its greatest asset is data. It isn't. The most valuable resource any company has lives inside people—the instincts, hard-earned lessons, mental shortcuts, and situational judgement that only experience creates. This is tribal knowledge. And it's vanishing every day.
In this reflection, I argue that the biggest structural weakness in modern organizations is the failure to capture and compound the wisdom of their people. I'm introducing the concept of a "Persistent Live Mind" that we're building at Sentium AI, designed to preserve and amplify this invaluable human context.
Tribal knowledge never sits in a document or a dashboard. It lives in neural pathways, surfacing only in conversations, crisis moments, or chance meetings. When a veteran leaves, burns out, or simply moves to another team, decades of wisdom disappear with them.
Companies pretend their systems hold their knowledge. They don't. Their people do—and there is no infrastructure to extract, preserve, or compound it. This is one of the biggest structural weaknesses in modern organizations.
I believe if tribal knowledge could be captured continuously and easily, the impact would be enormous. Knowledge compounds—what one person refines becomes the foundation for the next, and the next, and the next. But because tribal knowledge is undocumented and ephemeral, this compounding effect never happens. The organization is forced to "relearn" the same truths over and over.
At Sentium AI, we believe the next frontier isn't just automation—it's organizational memory engineering. We are building a Persistent Live Mind for enterprises: a voice-first intelligence layer that conversationally extracts tribal knowledge, structures it automatically, blends it with company documents and systems, applies real-time reasoning, and compounds this intelligence every day.
This isn't replacing people—it's preserving the best of them. The last decade was about capturing data. The next decade is about capturing wisdom.
Agentic Memory: The Next Frontier
Something profound has shifted. We've crossed into the era where prompts alone are no longer enough. The real frontier — the one that will define the next decade of AI — is agentic memory. I'm talking about contextual, persistent, synaptic, enterprise-grade memory — the kind of memory that begins to resemble the human mental stack.
Here, I outline my blueprint for a more advanced AI memory architecture, drawing parallels to human cognition (Memory Framework, Long-Term, Short-Term, and Episodic Memory). I make the case that memory is the true source of power in intelligent systems, and that building this capability is the core mission of Sentium AI.
Whether biology or machine, intelligence relies on memory architectures. Human cognition broadly operates across four layers: Memory Framework (our goals, principles, ambitions), Long-Term Memory (deeply stored knowledge: patterns, expertise, history, strategy), Short-Term Memory (the active working set—what just happened, what matters right now), and Episodic Memory (episodes, incidents, phases… the contextual richness that makes us human). And inevitably, there is Memory Atrophy—the fading of the irrelevant.
This is the blueprint of real intelligence. And now—for the first time—AI is beginning to replicate these layers.
As we build synthetic business consciousness for modern enterprises, we've learned one lesson repeatedly: Context is power. Memory is context. Therefore, memory is power.
Building in the Arena: The Realities of Startup Life
I provide an unfiltered look at the competitive realities of the tech industry. I champion the builders, defend the innovators, and offer commentary on the fast-paced, high-stakes environment of the AI landscape. My notes are a masterclass in resilience, focus, and maintaining integrity while navigating a world of intense competition and public scrutiny.
In Defense of the Brave
I'm no activist — far from it... But I want to take a moment to celebrate two young entrepreneurs from semi-rural India who have done something extraordinary... Disregard the noise. There is no signal there. just barely masked sour grapes! I say to GIGA AI, as I say to myself and our team - Its always hardest for the bravest!! Eyes on the prize!
In a powerful defense of Giga AI, I push back against industry criticism and celebrate the courage of fellow founders. This post highlights my belief in supporting innovators and focusing on building, rather than getting distracted by negativity.
The AI Olympics
The AI space has become incredibly competitive - like an Olympics where every player is bringing their A-game.
This post captures the intensity of the current AI landscape, comparing it to the Olympics. It reflects the immense pressure and exhilarating pace of innovation that defines the industry today, acknowledging the contributions of peers and the collective drive to push the boundaries of what's possible.
The Price of "Free"
MICE DIE IN MOUSE TRAPS BECAUSE THEY DO NOT UNDERSTAND WHY THE CHEESE IS FREE.
This simple but powerful phrase serves as a cautionary tale for me as an entrepreneur. It's a reminder to always question underlying motives, understand hidden costs, and look beyond the surface of what seems too good to be true—a critical lesson in both business and technology.
Voice, Ambient AI, and the Future of Human-Computer Interaction
At the core of my vision is a fundamental shift in how we interact with technology. I believe the future is not on our screens, but in the ambient environment around us, orchestrated by voice. My posts explore the implications of this shift, from the obsolescence of current devices to the creation of a truly seamless and intuitive user experience.
The End of the Phone, App, and OS
Elon just said - the phone, the app, and the OS will all die. 😳 All gone!!!
Reacting to a statement by Elon Musk, I underscore the coming paradigm shift in user interfaces. This note reinforces my conviction that voice-first, ambient AI will replace the screen-based interactions that have dominated the last two decades of technology.
The USB-C Moment of AI
⚡️The USB-C Moment of AI Has Arrived. The era of interoperability in AI has begun. Just like USB-C unified device connectivity, we're seeing the emergence of standards that will allow different AI systems, models, and agents to work together seamlessly.
Using the analogy of USB-C, I explain the importance of standardization in the AI ecosystem. I point to the Model Context Protocol (MCP) as a key enabler of this new era of interoperability, where different AI agents can communicate and collaborate, unlocking new levels of capability.
The Human Side of Technology: Purpose, Impact, and Ethics
For me, technology is not just about innovation for its own sake; it's a tool for profound social impact. Having witnessed poverty firsthand in India, I am driven by a deep-seated mission to build ethical AI that can create a more equitable and prosperous world. My notes often touch on the moral responsibility that comes with building world-changing technology.
A Mission Beyond Technology
Beyond technology, my mission is deeply human. Having worked across India, I've seen the moral failure of poverty amidst abundance. I believe ethical AGI, guided by empathy and purpose, can enable universal basic income and shared prosperity within our lifetime.
In my LinkedIn bio, I articulate a powerful mission statement that connects my work in AI to a larger social purpose. This highlights my commitment to building technology that serves humanity and addresses some of the world's most pressing challenges.
Industry Trends and the Pace of Change
I maintain a keen eye on the broader technology landscape, offering commentary on major shifts, announcements, and competitive dynamics. My notes provide valuable context for understanding where the industry is heading and what forces are shaping the future of AI and business.
The OpenAI Effect
(DON'T) RUN…THEY ARE COMING !!! Every time OpenAI makes an announcement, you can almost feel it…
This post captures the palpable tension and excitement that ripples through the AI community whenever a major player like OpenAI makes a move. It reflects the rapid pace of innovation and the constant pressure on startups to stay ahead of the curve while established giants continue to push boundaries.
Capacity Ahead of Demand
For as long as I can remember .. the masterful advice to us all .."Never put capacity ahead of demand" .. fast forward to 2025 ... a Billion is the new trillion ...and we are about $276 Billion ahead of demand !!!!!
I offer a sharp critique of the massive infrastructure build-out in AI, noting that the industry has invested hundreds of billions of dollars in data centers and compute capacity before demand has materialized. This challenges traditional business wisdom and raises questions about the sustainability of the current AI boom.
The Sentium AI Vision: Building MAXX and Composite Intelligence
My posts frequently return to the core mission of Sentium AI: building MAXX, the world's first truly autonomous, voice-first business partner. These notes provide insight into the company's unique approach, the problems we're solving, and the ambitious vision we're pursuing.
Meet MAXX: The AI Co-Founder
⚡️Our flagship product, MAXX, is the world's first truly autonomous, voice-first business partner and co-founder — is designed to remember, reason, speak, and act with contextual intelligence. Not just reactive, but proactive. Not just smart, but situationally aware. A new kind of AI — one that works like a team, sounds like a partner, and never sleeps.
This description of MAXX encapsulates the core value proposition of Sentium AI. It's not just another chatbot or assistant; it's an AI that understands context, maintains memory, and operates with a level of agency that makes it a true business partner.
Composite Intelligence: The Holy Grail
We believe the next stage of AI isn't about training bigger models — it's about capturing what no model has, the knowledge that lives inside us. The intuition, experience, and shared understanding that teams build over time — what we call tribal knowledge. Sentium is pioneering ways to ingest that human context and translate it into living, evolving intelligence.
This post articulates the central philosophy behind Sentium AI. Rather than competing on model size or compute power, we're focused on creating a new category of intelligence that blends the best of human wisdom with machine capabilities. This is the essence of Composite Intelligence.
Synthetic Business Consciousness
Say hello to Synthetic Business Consciousness at Sentium AI!
This bold statement signals a new frontier in enterprise AI. I envision a future where organizations have a persistent, evolving intelligence layer that captures, organizes, and amplifies the collective knowledge of their teams. It's a vision that goes beyond automation to fundamentally reimagine how businesses think, learn, and grow.
Lessons from the Road: Wisdom for Builders
Throughout my posts, I share hard-won lessons and principles that guide my approach to building companies, leading teams, and navigating the challenges of entrepreneurship. These notes are particularly valuable for aspiring founders and anyone seeking to understand what it takes to succeed in the high-stakes world of technology startups.
Stay in Your Lane, But Speak Up When It Matters
I'm a firm believer in staying in my own lane, minding my own business, and keeping my eyes locked on the prize. But... I want to take a moment to celebrate two young entrepreneurs from semi-rural India who have done something extraordinary.
I demonstrate the balance between focus and community. While I emphasize the importance of staying focused on your own work, I also show that there are moments when it's important to stand up for fellow builders and push back against unfair criticism.
The Courage to Build
It's always hardest for the bravest!! Eyes on the prize!
This simple mantra encapsulates the resilience required to build something truly ambitious. It's a reminder that the most innovative and transformative work often comes with the greatest challenges, and that courage and persistence are essential qualities for any founder.
Pray for the Day Your Idols Become Your Peers
As they say, pray for the day when your idols become your peers 😇.
This aspirational statement reflects my journey from outsider to insider, from admirer to peer. It's a powerful reminder that the people we look up to were once where we are now, and that with dedication and hard work, we can reach the same heights.
Context Engineering and the Architecture of Intelligence
My technical insights reveal a deep understanding of the underlying principles that make AI systems truly intelligent. My notes on context engineering, memory architecture, and the design of agentic systems provide a window into the sophisticated thinking that drives Sentium AI's approach.
Context is Power
Context is power. Memory is context. Therefore, memory is power.
This elegant formulation captures the essence of what makes an AI system truly useful. Without the ability to maintain and leverage context over time, AI remains limited to reactive, stateless interactions. By prioritizing memory and context, Sentium AI is building systems that can operate with a level of sophistication that approaches human intelligence.
The Five Layers of Memory
Human cognition broadly operates across four layers: Memory Framework, Long-Term Memory, Short-Term Memory, Episodic Memory... And inevitably, there is Memory Atrophy.
By mapping out the architecture of human memory and applying it to AI systems, I provide a blueprint for building more intelligent agents. This framework is central to Sentium AI's approach and represents a significant departure from traditional AI design.
The Model Context Protocol
The USB-C Moment of AI Has Arrived... we're seeing the emergence of standards that will allow different AI systems, models, and agents to work together seamlessly.
I recognize the importance of interoperability in the AI ecosystem. By embracing standards like the Model Context Protocol, Sentium AI is positioning itself to work within a broader network of AI tools and services, rather than building in isolation.
Personal Reflections and Gratitude
Beyond the technical and strategic insights, my posts reveal a deeply human side. I share moments of vulnerability, gratitude, and wonder, reminding everyone that behind every ambitious venture is a person navigating their own journey of growth and self-discovery.
The Intensity of the Journey
It's been an intense year for us at Sentium AI — the kind that stretches you, humbles you, and forces you to grow in ways you didn't plan.
This honest reflection acknowledges the personal cost of building a startup. It's a reminder that entrepreneurship is not just about strategy and execution, but also about personal transformation and the willingness to be stretched beyond your comfort zone.
The Wonder of Creation
Just so incredible to be back at UC Berkeley... flashback overload at the standahl syndrome ...all at the same time!
This note captures the sense of wonder and nostalgia that comes with reflecting on one's journey. It's a reminder to pause, appreciate how far we've come, and draw inspiration from the places and experiences that shaped us.
Conclusion: The Road Ahead
My "Notes from the Road" offer a glimpse into my thinking as I operate at the cutting edge of AI and human-computer interaction. My insights span the technical, the strategic, the personal, and the philosophical, providing a holistic view of what it takes to build something truly transformative.
At the heart of my work is a belief that the future of AI is not just about bigger models or more compute power, but about creating systems that understand us, remember us, and work alongside us as true partners. Through Sentium AI and MAXX, I'm pioneering a new category of intelligence—one that blends human wisdom with machine capabilities to create something greater than the sum of its parts.
For prospective investors, partners, and collaborators, these notes reveal who I am as a founder: deeply thoughtful, mission-driven, and committed to building technology that serves humanity. My journey is far from over, and the road ahead promises to be as challenging and exciting as the path I've already traveled.