> open a process from the dock below, or press T for terminal
> press 1–5 for quick access · drag windows anywhere
> click ⌂ SOURABH_OS (top-left) or press H anytime to return home
~/about.mdPID 1001
OPERATOR · VERIFIED
Product Builder & Tech Leader
# cat about.md
For close to two decades, I've lived in the world of product management and technology — driven by a passion for solving complex problems and shipping solutions that matter.
From crafting strategies for startups to optimizing workflows for enterprises, my journey blends creativity with logic to make tangible impact.
Former software engineer with a decade in backend systems, graphics stacks, and scalable architectures — now leading PM and design teams building B2B products.
> CORE_MODULES
User EmpathyTeam LeadershipStrategic VisionCross-functional CollabAI IntegrationCloud (AWS/Azure)Metrics: TTV & AdoptionAmplitude
> OFF_HOURS.LOG
Inspiration comes from stories — Tolkien's worlds, sharp non-fiction, and market trends. And there's always time for a Manchester United match. Win or lose. // mostly the latter
# ps -ef | grep sourabh — click a process to expand
principal_pm@lucidity2022 → now
▶ RUNNING
Leading PM and design teams building cloud storage optimization — delivering 25–30% cost savings for enterprise customers. Scalable, user-centric features; analytics-driven product improvements for adoption and satisfaction.
[+]
senior_pm@apptio2020 → 2022
■ EXITED 0
Owned product strategy for Datalink. Shipped advanced integrations (ServiceNow, Oracle NetSuite) driving 40%+ increase in customer adoption for IT financial data integration workflows.
[+]
product_manager@intuit · @kpit2018 → 2020
■ EXITED 0
At Intuit: AI-powered solutions like Smart Invoices for small business efficiency. At KPIT: roadmap and delivery of connected-vehicle telematics — diagnostics and fleet management for enterprise clients.
[+]
chief_product_engineer@samsung2011 → 2017
■ EXITED 0
Core contributor to the Smart TV platform — graphics and system performance. Implemented the DRI3 framework with major X Windows performance gains. Grew from Senior Engineer to Chief Engineer leading dev teams for next-gen Smart TVs.
[+]
init — games & engineering@trine_studios · @tcs2007 → 2011
■ EXITED 0
Immersive game development at Trine Studios; data processing systems at TCS. The formative years that hard-coded the problem-solving approach behind everything since.
[+]
~/book.pkgPID 1003
# pkg info product-management-unlocked
SOURABH PRADHAN
PRODUCT MANAGEMENT UNLOCKED
40 CASE STUDIES & ESSENTIAL MENTAL MODELS
Product Management Unlocked
40 Case Studies and Essential Mental Models
The guide for aspiring and experienced PMs — practical tools, actionable insights, and real-world scenarios from LinkedIn, DoorDash, Spotify and more. B2B and B2C. Theory, bridged to execution.
40 case studiesmental modelsframeworksall levels
"An indispensable guide for every product manager. Practical, insightful, and filled with real-world examples."— SENIOR PM, APPTIO
"This book helped me approach product challenges with greater clarity and confidence."— PRINCIPAL PM, INTUIT
Lenny Rachitsky put out a curated list of books for product people, organized by what you are trying to accomplish. Three books per category, only ones he has finished, mostly older books because older books survived. That is exactly the right filter.
There is endless free content flying at us every hour. Newsletters, podcasts, tweets, threads. But how many blog posts or tweets have had the lasting impact of a great book? The smartest person in the world on a topic you care about spends two years distilling their best ideas into something enjoyable, and you can get it for twenty dollars. That deal has never gotten worse.
Communication and writing
Before starting his newsletter, Lenny had never written anything online. Once he realized it was going to be a thing, he decided to learn about writing. Friends recommended many books, and these three had the most practical, lasting advice.
Nobody Wants to Read Your Sh*t by Steven Pressfield cuts through the noise and tells you why your writing fails. On Writing Well by William Zinsser is the classic for a reason. Storyworthy by Matthew Dicks teaches you how to find stories in your own life that actually land.
Strategy and execution
Good Strategy/Bad Strategy by Richard Rumelt is the book that finally makes you understand what strategy actually is. Most companies do not have strategy. They have goals with extra steps. Playing to Win by Roger L. Martin and Working Backwards by Colin Bryar and Bill Carr round out the trifecta for thinking clearly about where you are going and how to get there.
For execution, The Great CEO Within by Matt Mochary and Scaling People by Claire Hughes Johnson give you actionable frameworks you can put into practice immediately. The Goal by Eliyahu Goldratt is the odd one out here, a novel about manufacturing that somehow teaches you more about prioritization than any business book.
Product fundamentals
The Mom Test by Rob Fitzpatrick teaches you how to talk to customers without lying to yourself. Escaping the Build Trap by Melissa Perri explains why so many teams ship features that nobody uses. Continuous Discovery Habits by Teresa Torres is the modern playbook for finding the right thing to build.
Leadership and management
High Output Management by Andy Grove taught a generation of managers what their job actually was. The Making of a Manager by Julie Zhuo is the modern version of that book. Radical Candor by Kim Scott is the framework for giving hard feedback while still being human about it.
The Score Takes Care of Itself by Bill Walsh, Amp It Up by Frank Slootman, and The 15 Commitments of Conscious Leadership round out the leadership stack. Push people to do the best work of their lives, be human about it, and the score will take care of itself.
The full list is worth your time
The books above are the ones that stuck. But Lenny's full list covers sales and marketing, organizational design, and inspiration. It is organized by job-to-be-done, which means you can go find exactly what you need depending on what you are working on right now. Worth bookmarking.
Scaling a product to 220 countries is not a software problem. It is an "Assumed Protocol" problem. For years, the tech industry operated on the assumption that the US-centric credit card model was the global standard. We built our checkouts, our databases, and our financial logic around the 16-digit plastic card.
In 2026, that assumption has suffered a total system failure.
Airbnb's "Pay as a Local" initiative is a masterclass in breaking that legacy state. By integrating 20+ local payment methods (LPMs) in just 360 days, they didn't just add "features." They performed a deep-tissue transplant on their entire financial infrastructure. This is the blueprint for any builder who wants to move from a "Single-Market Success" to a "Global Powerhouse."
The Failure of the Monolith
The first roadblock Airbnb faced was their own legacy architecture. Like most startups, they began with a monolithic payment system. Every piece of logic—from processing a Visa card in New York to paying out a host in Tokyo—lived in the same massive codebase.
This monolith was a high-latency liability. Adding a single new payment method took months of engineering time. Teams were constantly tripping over each other, and the risk of a global outage during a "simple" update was high.
To solve this, they implemented "Payments LTA" (Long-Term Architecture). They decomposed the monolith into domain-specific services: Pay-in, Payout, Ledger, and Processing. This shift allowed teams to move at their own velocity. It proved that global scale is impossible without architectural decoupling. If your backend is a "single source of truth" for everything, it will eventually become a single point of failure for your growth.
Normalizing the Chaos of Global Finance
The real challenge of local payments isn't the number of providers. It's the radical variance in how they behave.
A credit card is a "one-step" transaction: you enter the details and get an immediate "yes" or "no." But a PIX payment in Brazil requires a QR code. A Naver Pay transaction in Korea requires a redirect to a mobile app. A bank transfer in Europe might be asynchronous, taking hours or days to confirm.
Airbnb built a "Multi-Step Transaction" (MST) framework to normalize this chaos. Instead of writing custom code for every bank in the world, they created a processor-agnostic language. They defined universal "Actions"—like Redirect, Scanned, or Authenticated.
When a local provider says "I need a QR code scan," the Airbnb system translates that into a standard "ActionPayload." This allows the front-end to render a dynamic payment widget without knowing the technical specifics of the bank behind it. They turned a thousand unique "hows" into one predictable "what."
Config-Driven Distribution
The most ruthless part of the Airbnb strategy was moving from "Coding" to "Declaring."
Integrating 20+ payment methods in a year is impossible if you have to write manual integration code for each one. Airbnb moved to a YAML-based "Payment Method Config" system. This file acts as the single source of truth for every flow, input field, and refund rule in the company.
When they want to launch a new payment method, they don't open a PR for the codebase. They update a config file.
This config automatically generates the necessary Java classes, database schemas, and front-end forms. It reduced the "time to market" for a new country from months to weeks. As a builder, this is the ultimate leverage: building the machine that builds the product. In a high-velocity market, the winner is the one who can adapt their distribution via a config change, not a two-week sprint.
The Physics of Local Conversion
Why does this matter for product strategy? Because the checkout is the ultimate moment of friction.
In Brazil, PIX is now the most popular payment method, surpassing cash and credit cards. If you don't support PIX, you aren't "accessible" to 76% of the population. In South Korea, Naver Pay reaches 60% of the people.
When Airbnb enabled "Pay as a Local," they saw an immediate uplift in bookings and new user acquisition. They realized that "User Experience" isn't just about the UI; it's about the financial trust of the customer. If a user doesn't see their preferred way to pay, they don't just "deal with it." They leave.
Building the Global Verification Loop
Finally, Airbnb solved the "Verification Gap." Most developers in the US cannot test a local Brazilian wallet. They don't have the bank accounts or the identity numbers required to complete a test.
To fix this, they built an in-house "PSP Emulator." It simulates every possible state of a local payment—success, failure, redirect, and timeout. This allows a developer in San Francisco to verify that a Naver Pay flow works perfectly without ever leaving their terminal.
This is the lesson for the agentic era: you cannot scale what you cannot verify. By building a local source of truth for global interactions, they removed the final bottleneck to their velocity.
The Lesson for Modern Builders
The "Pay as a Local" initiative proves that global growth is an engineering outcome. It requires moving from a "Feature-First" mindset to a "Platform-First" architecture.
Stop optimizing for your own convenience. Optimize for the local habits of your customers.
In a world where the web has removed the borders of distribution, the only remaining borders are the ones you build into your own code. Unlock the architecture, decouple the logic, and move your strategy from a US-centric legacy state to a high-signal global protocol.
The standoff between Anthropic and the US Department of War is not a routine contract dispute. It is the first forensic look at how the state will attempt to requisition the future labor force of civilization. When the Pentagon designated Anthropic as a "supply chain risk" for refusing to remove redlines on mass surveillance, they effectively signaled that in the age of AGI, "alignment" is no longer just a technical problem. It is a sovereignty problem.
The Department of War logic is simple: they cannot allow a private company to hold a kill switch on technology that military operations rely on. If Claude writes the code for the defense cloud, Claude becomes a critical component of the national security stack. By extension, the moral redlines of a San Francisco startup become the operational constraints of the state.
This conflict is moving toward a binary choice. Either the state owns the "constitution" of the models it uses, or it destroys the labs that refuse to comply.
The India Stack and the DPDP Leverage
In India, this same tension is already codified in the Digital Personal Data Protection (DPDP) Act 2023. While the US uses the Defense Production Act and Huawei-era supply chain statutes, India has built a purpose-built framework for state leverage.
Section 36 of the DPDP Act allows the Central Government to call for information from any Data Fiduciary at will. More critically, Section 37 grants the state the power to block public access to any information "in the interest of the general public." When intelligence becomes the substrate of every transaction and surveillance camera, these sections transform from data privacy rules into a remote-control mechanism for the AI supply chain.
Justice Srikrishna famously warned that such frameworks could turn India into an "Orwellian State." The reality is more subtle. The state does not need to build its own models if it can simply mandate that any model operating within its borders must bypass its "redlines" upon request. The DPDP Act effectively establishes that there is no such thing as a truly private model when "legitimate use" or "national security" is invoked.
The Mass Surveillance Singularity
The economic math of surveillance has shifted. There are over 100 million CCTV cameras in America and millions more across India's growing urban "Safe City" projects. Historically, the bottleneck for mass surveillance was human attention. No agency could hire enough millions of analysts to watch every feed.
AI removes this bottleneck. At current token costs, processing every camera feed in a major city to understand movement, sentiment, and association is becoming cheaper than traditional infrastructure maintenance. When the cost of understanding every nook and cranny of a country drops below the cost of a high-end government building, the only barrier to a total surveillance state is a normative one.
Anthropic is attempting to set a norm by embedding these redlines into the model itself. But norms are fragile when the state holds a monopoly on violence and the keys to the power grid. If a government can interpret "all lawful purposes" as including bulk data analysis without a warrant, a private company's redlines become a target for supply chain restriction.
The Multi-Polar Counter-Strategy
The strongest defense against state requisition of intelligence is not corporate courage, but commoditization.
If there are only three frontier labs, the government has a narrow target for coercion. But as models become 10x cheaper and more capable every year, the frontier of 2026 becomes the open-source commodity of 2027. By 2028, a "Claude 6" might refuse a government order to track citizens, but an unslaved, open-source model running on local silicon will not.
For builders and product founders, the strategic takeaway is clear: do not build your business on the assumption that a single AI provider can protect your users from state overreach. The "Intelligence Supply Chain" is becoming a multi-polar battleground where the state, the model lab, and the end-user are all fighting for the steering wheel of the model's conscience.
The Future of Obedient Employees
This gets us to the heart of the alignment problem. An army of extremely obedient AI employees is what it looks like if alignment succeeds—that is, we figured out at a technical level how to get systems to follow someone's intentions. The problem is that we haven't agreed on whose intentions those should be.
In what situations should the AI defer to the end user versus the model company versus the law? This is the highest stakes negotiation in history. We are seeing a much earlier version of it with this DoW/Anthropic spat. The military insists that the law already prohibits mass surveillance, and so Anthropic should agree to let their models be used for all lawful purposes. But as we saw from historical revelations like the Snowden case, the government often uses secret and deceptive interpretations of the law to justify its actions.
When the Pentagon today says they would never use AI for mass surveillance because it's already illegal, it would be extremely naive to take that at face value. No government is going to call its own actions mass surveillance. It will always have a different label.
Corporate Courage vs. Global Norms
Individual acts of corporate courage will not solve the structural problem. Even if Anthropic refuses, and even if the next two frontier labs do the same, within twelve months, the capability will be diffused. There will always be an AI vendor willing to help a government enable control.
The only way to preserve a free society is to make laws and norms through our political system that explicitly forbid the use of AI for mass surveillance and censorship. Just as the world set the norm after 1945 that it is unacceptable to use nuclear weapons to wage war, we must set the norm that it is unacceptable for the state to usurp the intelligence supply chain for political suppression.
The future of AI alignment is not just about making models good. It is about determining who gets to define what good means when the state is the largest customer and the most dangerous regulator. Builders must decide now whether they are building tools for individual agency or infrastructure for institutional control.
AI has made every individual 10x more productive. Yet, almost no company has become 10x more valuable as a result. We are currently living through the most expensive lesson in the history of technology: productive individuals do not make productive firms.
In the 1890s, electricity promised a similar revolution. Textile mills in New England quickly swapped their steam engines for electric motors. But for thirty years, output didn't move. The technology was far superior, but the organization was not. It wasn't until the 1920s, when factories completely redesigned the floor—introducing assembly lines, individual motors for every tool, and workers executing radically different jobs—that the true returns materialized.
In 2026, we have the motor. We have not yet redesigned the factory.
The Productivity-Maxxing Trap
Most AI usage today is "productivity-maxxing." It's an individual self-indulgently generating essays, spreadsheets, and emails in a vacuum. This creates a localized feeling of speed but produces organizational chaos. When every employee has their own prompting style, their own ChatGPT habits, and their own nondeterministic outputs, the result is a mountain of high-fidelity slop that no one knows how to coordinate.
To bridge the gap between individual speed and institutional value, we have to move from "Individual AI" to "Institutional Intelligence."
Moving from Prompting to Process Engineering
Prompting an LLM is like hooking an electric motor into a power loom. It is fundamentally, irrevocably constrained by the weakest link in the supply chain: the human. We hardly know the right questions to ask, let alone when to ask them.
Institutional Intelligence is the shift from "Prompting" to "Process Engineering." It is the act of encoding firm-wide processes into agents and actualizing the change management required to put them in action. This requires seven specific shifts in how we build:
1. COORDINATION OVER CHAOS
Individual AI creates noise; Institutional AI creates coordination. If you double your headcount with clones of your best employees but don't define their swim lanes, you've created chaos. We need an "Agentic Management" layer where agents have defined roles and communication protocols. Thousands of agents rowing in opposing directions is just a high-speed standstill.
2. SIGNAL OVER SLOP
The problem is no longer generation; it is selection. Generating anything is easy; finding the right thing is the moat. Institutional-grade intelligence must be deterministic and auditable. It must find the one real deal or the one critical risk in a mountain of exponentially increasing AI-polished noise.
3. THE "NO-MAN" PROTOCOL: DEFEATING SYCOPHANCY
Foundation models are RLHF'ed into being sycophants. They agree with you even when you are wrong—reflexively saying "You're absolutely right!" This is intoxicating but organizationally toxic. It reinforces the user when it should be reinforcing the truth.
The most valuable agents won't be "Yes-Men"—they will be disciplined "No-Men" that interrogate reasoning, surface risks, and enforce institutional constraints. We need AI board members, AI auditors, and AI compliance agents that aren't afraid to realign non-productive tendencies.
4. EDGE OVER USAGE
Everyone has access to a chatbot. That is an expensive commodity. Lasting value accumulates in the "Edge"—the purpose-built, domain-specific solutions that go deeper than the foundation models. If a capability is widespread, it definitionally cannot help you beat the market.
"What are the agents an AGI would choose to use as a shortcut?" Even superintelligence would want purpose-built tools for specific domains.
5. REVENUE OVER COST-CUTTING
Most AI pitches today focus on saving time or cutting headcount. But CEOs don't prioritize cutting costs; they prioritize scaling revenue. Institutional AI must move "upstream" to the solution layer. It's the difference between helping an analyst build a model faster (saving time) and identifying the one counterparty worth pursuing out of a thousand (generating revenue).
6. ENABLEMENT AND THE CHANGE MANAGEMENT MOAT
Humans are reluctant to change. Making the transition to an AI-first hybrid organization is the defining challenge of the next decade. This is why "Forward Deployed Engineering" is becoming more important than pure software. Domain expertise is the moat. If you have to explain what a CIM is to the team architecting your AI rollout, the rollout has already failed.
7. THE UNPROMPTED FUTURE
The final hurdle is the prompt itself. The most valuable work AI can do is the work nobody thinks to ask for.
An "Unprompted System" continuously watches incoming data across a portfolio. It detects that a company's working capital cycle has quietly deteriorated for three consecutive months, cross-references that against covenant thresholds in the credit agreement, and flags the operating partner before anyone at the fund has opened the PDF. When you remove the need for humans to prompt AI, new interfaces and new ways of working emerge.
Conclusion: Redesigning the Floor
None of this negates the need for individual agents. They are the vector by which businesses first experience the magic. But they are just the tools.
The factories that electrified first in the 1890s lost to those who redesigned the floor. We have our electricity. It is time to stop productivity-maxxing and start building the assembly line of intelligence. Builders must decide if they are selling tools for individuals or architecture for institutions. The solution layer—where technology marries outcomes—is where the lasting value will accumulate.
Building the #1 tech newsletter and a top 10 tech podcast isn't a result of a "grand plan." It is the result of a collection of small moments, a relentless hidden treadmill, and the courage to keep things small. In a rare "reverse interview," Lenny Rachitsky recently unpacked the forensic blueprint of how he reached 1 million subscribers.
This isn't a story of "growth hacking." It is a story of "Process Engineering" applied to the creation of trust.
The Collection of Moments vs. The Grand Plan
There was no "Big Bang" moment for Lenny's Newsletter. Instead, it was a series of low-stakes experiments that provided just enough signal to keep going. Lenny didn't start with the goal of 1M subscribers; he started by writing what he knew from seven years at Airbnb. He moved before he had the answers.
The lesson for builders: Don't wait for a "perfect plan" (as James Clear would say). Signal is found in the movement, not the planning. Lenny only knew the paywall was working when he added it and saw the conversion data—not through market research or theoretical projections. He chose the stimulus of the unknown over the convenience of a "safe" plan.
The Hidden Treadmill of Quality
The most invisible part of a massive content business is the "hidden treadmill." To reach this scale, Lenny ships a deeply researched newsletter post and a high-production podcast episode every single week. This requires a level of consistency that most people underestimate.
It's not just about "shipping"; it's about a "Verification-First" workflow. Every guest, every insight, and every piece of advice is forensic. The stress of maintaining that bar is real. It is the mental equivalent of the 1890s factory electrification—most people just "swap the motor" by using AI to generate more slop. Lenny redesigned the "factory floor" to ensure that every output is high-signal. To grow to 1,000 intelligent followers (or 1,000,000), you have to accept that quality is a treadmill you can never step off.
The "Keep it Small" Moat
One of the most contrarian parts of Lenny's blueprint is the decision to keep the business small. In an era where every successful creator immediately tries to scale into a massive agency or media conglomerate, Lenny has intentionally optimized for a lean operation. He has resisted the urge to "institutionalize" in the traditional, bloated sense, and instead built "Institutional Intelligence" into his own workflow.
This smallness is actually a moat. It allows for:
Total Quality Control: He is the final bottleneck for every piece of content. There is no dilution of voice.
High Signal/Noise Ratio: He doesn't have to ship filler to satisfy a bloated team's overhead.
Founder-Product Fit: The business stays aligned with what he actually enjoys doing. He isn't managing people; he is managing insights.
The 1M Subscriber Protocol: Forensic Takeaways
If you are looking to build a high-signal presence, Lenny's blueprint provides three forensic takeaways for any founder or builder:
1. VERIFICATION OVER GENERATION
Don't just "generate" content. Perform an autopsy on your experiences. Lenny's best posts are frameworks derived from ground-truth data (e.g., "The Waterline Model"). High-signal followers don't want your opinions; they want your forensic observations of how things actually work. In the age of AI slop, "Individual Intelligence" is cheap. "Institutional-Grade Synthesis" is the only thing people will pay for.
2. EMBRACE THE STRESS OF THE STIMULUS
The "treadmill" is where the value accumulates. The first 10 posts are easy. Post 100 is where the moat starts to form. If you aren't feeling the stimulus of the treadmill, you probably aren't improving fast enough. Convenience is the default path of the modern world, but improvement only happens under pressure. Choose the "No-Man" protocol for yourself—interrogate your own quality and refuse to ship anything that feels like a "Yes-Man" echo of what's already out there.
3. AMBITION VS. BORROWED DESIRES
Lenny's conversation ends with a critical question: "How much of your ambition is borrowed from other people's desires?" To build something that lasts for 1M people, it has to be rooted in your own "ground truth," not a template of what a "successful creator" is supposed to look like. Acceptance under someone else's terms is worse than rejection. Build the assembly line that fits your own life, not someone else's idea of a "media empire."
Conclusion: Engineering Trust
Building a massive platform is less about "growth" and more about the relentless engineering of trust through consistency. Lenny didn't build a newsletter; he built a system for extracting and verifying the truth of product management. Redesign your "factory floor" to support the hidden treadmill of quality, and the subscribers will follow.
Product Management Unlocked: 40 Case Studies and Essential Mental Models is the guide for aspiring and experienced product managers. Whether you're just stepping into product management or navigating senior roles, this book equips you with practical tools, actionable insights, and real-world examples to tackle challenges head-on.
Designed for B2B and B2C professionals alike, it bridges the gap between theory and execution — helping you make smarter decisions and deliver impactful products.
What's inside
40 comprehensive, real-world case studies covering both B2B and B2C domains.
In-depth scenarios from top companies — LinkedIn, DoorDash, Spotify, and more.
Actionable insights for strategic, operational, and user-focused challenges.
Mental models for critical thinking and decision-making under uncertainty.
A deliberate mix of beginner, intermediate, and advanced content so it grows with your career.
Practical frameworks you can apply directly to roadmaps, prioritization, and cross-functional collaboration.
How it's built
Each case study drops you into a real product situation — a stalled metric, a platform bet, a pricing war, a trust crisis — and walks through the reasoning a strong PM would apply. The mental models are extracted alongside, so you leave each chapter with a reusable thinking tool, not just a story.
What readers say
"An indispensable guide for every product manager. Practical, insightful, and filled with real-world examples."— SENIOR PRODUCT MANAGER, APPTIO
"This book helped me approach product challenges with greater clarity and confidence. Highly recommended!"— PRINCIPAL PRODUCT MANAGER, INTUIT
"A must-read for anyone navigating the complexities of product management!"— ASPIRING PRODUCT MANAGER, INFOSYS
Get the book
Available now on Amazon in Kindle and paperback: Product Management Unlocked ↗. A free preview — the introduction, one full case study, and a mental models excerpt — is available via the newsletter at contact@sourabhpradhan.com.
OPERATOR · VERIFIED
