It started with stories. 5th and 6th standard — deeply lost in Panchtantra, Sudha Murthy's books, and I.K. Vijaliwala's writing. English stories weren't homework, they were portals. The inclination towards Maths and English started right here — not from textbooks, but from the obsession with how words and numbers could create worlds.

Books, writing good essays, and having the best handwriting in class were the primary charms. Won 2nd prize in a literature award. By 7th-8th standard, the stage became a second home — anchoring school events became a constant. Writing small poetry. Another 2nd prize, this time in poetry.

Then sports took over. 8th to 10th standard was pure energy — Kabbadi became serious, reaching 3rd place at State Championship. But it wasn't just Kabbadi.

Kabbadi. Kho-Kho. Long Jump. 200 Meter Race. Cricket — five sports, all at competitive level. The discipline of showing up to practice every morning before school, the strategy of reading an opponent's body language — these weren't just games. They were the first lessons in reading systems.

Simultaneously, science started fascinating. Physics — how things actually work. Not the textbook version, but the real curiosity: why does this bridge hold? How does electricity actually flow? Multiple school projects around carbon emission and sustainability — long-term solutions that got recognised at state level.

"Before I could write code, I was already trying to decode every system I could find — on the field, in stories, in the way things worked."

11th and 12th standard changed everything. The sports energy channelled into pure academic intensity. Not just studying — mastering. The kind of obsession where you solve a physics problem and then solve it two more ways to understand the underlying structure.

293 out of 300. Top 10 position in the entire state. 99 + 99 + 95. Not a fluke — the result of a discipline mindset, hard work meets smart work, and the genuine fascination with how equations describe reality.

This is where the real thinking started: mastering inertia, angles, ratios, distributions, integrations. Solving polynomial equations. Solving time series. Tackling real-world physics — speed problems with angle, friction, and inertia. Probability mastering. Permutation & combination mastering.

These weren't just exam topics. They were the vocabulary that would later let me think about trading algorithms, data distributions, and system optimisation — the foundation for everything that came after.

"293/300 wasn't a score. It was proof that obsession, channelled correctly, produces results that compound forever."

1st year was scouting — figuring out what engineering actually is. Taking part in multiple societies, being part of everything, trying to understand where the real learning happens. Spoiler: it wasn't in the lecture halls.

2nd year — the intensity shifted. Gaming. Motion graphics. Hardware knowledge. Real computer figuring out — not just using machines, but understanding them at the metal level. What makes a GPU different from a CPU. How memory works. The kind of knowledge that makes you dangerous with code later.

3rd year — Vice President, Technical Society. Organised the 1st ever Technical Fest of NIT Jaipur. A whole new concept — from planning to partnering with artists, creators, and tech people, to publicising and executing. An event built from zero.

4th year — Advisor to the Technical Society. And a private obsession that would change everything: entered into trading. Not casual investing — the deep end. Algo trading. Web scraping. Statistics. Options & Futures. Stocks fundamentals. Balance sheets. How companies form and operate.

While everyone else was preparing for placements, I was building scrapers that pulled financial data at 3am, backtesting strategies that nobody asked me to build, and learning more about markets than any course could teach.

"College didn't teach me engineering. It taught me how to teach myself anything."

JSW Steel (2019–2020) — leading 25+ people, 144 Coke Oven Batteries, shift rotations, yield targets. Robotics and automation in a high-risk, compliance-heavy production environment. Temperature and pressure sensitive processes — the kind where a small error isn't a bug, it's a safety incident.

Won the Best Kaizen Award. Worked on Six Sigma projects. Real production improvement — efficiency, automation, productivity at industrial scale. All about making systems do more with less. And in between the factory shifts: hosted Saina Nehwal, anchored company functions for 20,000+ people gatherings. The anchoring instinct from school days never left.

Comcast (2021) — from a factory floor in India to US Telecom. West zone, marketing and sales funnel. 6 books worth of domain knowledge absorbed. 1M+ datapoints. Customer data at a scale that changes how you think about analytics — not hundreds, but millions of data points flowing through the funnel.

Mastering marketing and sales funnel analytics at scale for a US Telecom giant. Managing leadership-level clients. This was the phase where data stopped being a tool and became a thinking framework.

"JSW taught me how physical systems work. Comcast taught me how human systems buy. Both follow the same rules."

Spinny (2021–2023) — this wasn't a data job, it was an operating role. Defined 25+ variables, 20+ KPIs. Built dashboards that became the operating layer for the entire used car inventory business. Every decision — pricing, supply, demand, regional allocation — ran through systems that were built in this phase.

200+ agents. 5× growth in the top 8 markets. 61.5% cost reduction. These numbers weren't targets hit — they were the output of systems designed to scale themselves. The dashboards didn't just report performance, they drove it.

In between all this scale, there was BasicKart — a scrappy attempt at a startup as a partial founder. Building the product, struggling with the market, failing fast. It was the foundational reality check before returning to big systems.

Left JSW with Kaizen awards. Left Comcast with CX mastery. Left Spinny with the conviction that if you can define the right variables and build the right dashboard — any system can be operated at 10× with the same resources.

"At Spinny I learned the difference between analysing a business and operating one. It's the difference between reading a map and driving the car."

2023 started with agricultural supply chains. Real operations, real logistics. Not every system you build belongs to digital spaces — some involve real trucks, real weather, and real physical warehouses.

Agrim — B2B marketplace. Warehouse automation, EDD rule engine, A/B testing infrastructure. Took dispatch success from 89% to 96%. Built a 17L+ retailer lead database from zero. 10L+/month in reconciliation savings.

CredResolve (2024) — lending and fintech. First time everything converged. Full Product Lead — frontend, backend, both teams. Voice AI, NLP models, and the realisation that building a product end-to-end is fundamentally different from building features inside someone else's system.

"Every failure is just data for the next iteration. Full stack building means taking ownership of the entire system."

2025 was the year of building the AI sandbox before the business. The question was not "how do I call a model API faster?" It was "how do I create a private environment where models, tools, memory, and execution can collaborate without data leakage?"

A tight AI council/community became the proving ground. Every new idea was pressure-tested there first: prompt systems, agent skills, retrieval loops, cleanup pipelines, evaluation runs, and shared-memory patterns that could survive beyond a single chat window.

Opus 4.6 became the catalyst. That release opened up a much deeper way of working: model harnessing, structured tool calling, local memory stitching, and reusable private compute flows. Instead of building one-off demos, I was building an operating environment.

The experiments turned commercial fast. One AI product launched publicly on Product Hunt, and two heavier P2P product sales landed privately with buyers in Dubai and Norway. 2025 was not polished consulting yet. It was the lab where the consulting moat got invented.

"The sandbox mattered because it proved I could build intelligence systems that stay yours, remember context, and keep getting better without depending on somebody else's cloud."

2026 changed the mandate completely. The goal was no longer to just prove things inside the sandbox. It was to solve outside problems using the same private-AI discipline and install that capability inside other businesses.

ZeroOne D.O.T.S. AI Consulting became the external layer for everything learned before it. Instead of freelancing for tasks, the model became consulting through systems: understand the workflow, deploy private compute, connect the right tools, and create an environment where the client's own data can safely power the product.

This is where the stack became concrete and visible. 19 live subdomains. 24+ Docker services. OpenClaw running multi-channel AI gateways. VibeCode exploring what an agent-native IDE should feel like. Skills and commands published back into the ecosystem so the tooling itself could create compounding leverage.

The deeper shift was identity-level: not just builder, not just operator, not just product person. An independent systems architect shipping private AI environments for real clients.

"2026 is where the private AI lab became a consulting engine: same obsession with systems, but now installed directly into other people's operating environments."