About Me
Product Manager by profession. Entrepreneur by passion. Builder by nature.
I build products at the intersection of AI, enterprise software, and startup ventures. From AI-native PM platforms to production-grade observability systems, from enterprise process mining to e-commerce marketplaces — I've shipped across every layer of the stack.
What makes me different is that I don't just manage products — I build them. I use Claude Code as my primary development tool, shipping production-grade software at 10x the velocity of traditional development.
Product Philosophy
Ship Fast, Learn Faster
I believe in getting products to users quickly and iterating based on real feedback. Clarion went from idea to 30 users in 15 days — not because it was perfect, but because it solved a real problem.
Full-Stack Thinking
Great PMs understand the technology deeply enough to make hard tradeoffs. I've built backends in Node.js, Python, and Go — not to write production code, but to speak the language of my engineering teams fluently.
AI as a Force Multiplier
I don't just manage AI products — I build with AI. Claude Code is my primary development tool. The future of product management is deeply intertwined with AI, and I operate at that frontier daily.
Evidence Over Opinions
Every product decision should be grounded in data. From customer signal analysis in Clarion to process mining in Glaux, I build systems that turn raw feedback into prioritized action.
How I Build
AI-augmented engineering
Claude Code is my primary development tool. I ship production platforms solo at the velocity teams of 10+ engineers ship — Clarion at 119K lines of TypeScript in weeks, Strix at 234K. The leverage isn't theoretical; it compounds.
CLAUDE.md as the single source of truth
Context engineering matters more than prompting. When my AI assistant has the full picture — conventions, patterns, constraints, domain model — in one file, it produces production-quality code consistently. CLAUDE.md becomes the soul of the codebase.
Machine specs, not vague PRDs
"Make it fast" is the enemy of agent-native development. I write quantified specs — <200ms p95, layouts at 640/768/1024px, OWASP Top 10, bcrypt(12). Coding agents produce working code on the first attempt, not the fifth.
Closed-loop product process
Discovery → spec → ship → measure → recalibrate. Every decision traceable from a customer quote to a shipped feature. Scoring weights learn from outcomes — if predicted impact ≠ actual, the system tunes itself. This was the core thesis behind Clarion.
What I Believe
PMs who don't code are getting left behind
Deep technical fluency means better product decisions, faster iteration, and the ability to spot opportunities pure-play PMs miss. I've built five ventures from scratch precisely because I refuse to let engineering be a black box. The future of product management is intertwined with AI — PMs who can't operate at that frontier will be replaced by ones who can.
AI-native beats AI-bolted-on, every time
Datadog's LLM observability scores 6/10 because they're retrofitting prompt tracking into spans designed for HTTP requests. Strix scores 10/10 because I built LLM-native data models from day one. The same pattern plays out across categories — process mining (Glaux vs Celonis), PM tools (Clarion vs Productboard), code generation (Ultracode vs the iframe-preview tier). Architecture decided at day-one is the moat.
The full loop is the real moat. Point tools always lose
Every Clarion competitor solves 2-3 stages. None connect all seven — signals → intelligence → opportunities → specs → agents → validation → outcomes. Building the full pipeline as one platform is a structural advantage integrations can't replicate. Data flows natively between stages with shared identity, taxonomy, and scoring.
A single founder with AI tooling can outbuild a team of five without it
I shipped Clarion's 230+ features in 5 weeks solo. Strix at 234K LOC, architected end-to-end. AI removes the velocity bottleneck of human-only engineering — the rate-limiting step is no longer typing speed but architectural clarity. The competitive landscape is being rewritten by builders who know how to wield it.
80% of global enterprises are still locked out of the AI revolution
Most AI tooling assumes Silicon Valley resources — cloud-native, unbounded data egress, no data-residency constraints. The real market is F500 enterprises running legacy infrastructure under regulatory pressure. That's why Glaux is on-prem-capable, multi-LLM by default, and advised by the ex-COO of HP Enterprise. The next $100B in enterprise AI is built for them, not despite them.
What I'm Looking For
Open to conversations where the mission, the team, and the technology align. I'm selective — but the right opportunity is worth the move.
Roles that fit
- Founding Product / Founding PM at an AI-native enterprise startup
- Head of AI Product or VP Product at a Series B–D SaaS
- Founder / Co-founder for a venture combining technical depth and go-to-market
- Strategic AI advisor or operator role at a fund's portfolio company
Domains I work in
- AI tooling and developer infrastructure (Ultracode, Clarion territory)
- Enterprise observability and dev tools (Strix territory)
- Process intelligence, regulatory and compliance AI (Glaux territory)
- AI-augmented PM and design tools
Working style
- Small, focused teams that ship weekly, not quarterly
- AI-native engineering culture — Claude Code, MCP, agentic workflows
- Customer-facing involvement encouraged, not avoided
- Remote / hybrid (Gurgaon-based); open to relocation for the right mission
Not for me
- Pure consumer apps without a clear B2B angle
- Ad-tech, web3 speculation, or "AI-as-feature" wrappers
- Pure technical IC roles — I want product ownership
Beyond Work
Sir Syed Educational Society
7,000+ educated · 1,000+ skilled · 800+ employed
Ongoing community work in education and skills training. The most meaningful project I'm part of doesn't ship code. It's a long arc — multi-decade — and a humbling reminder that scale isn't always digital.
Yum Spices — family business turned platform
30-woman workforce · 20+ partner farms · +25% farmer income
My family has produced premium spices for generations. I built the D2C platform that brought our supply chain to Indian consumers — and along the way made the workforce all-women at 3× village wages, increasing partner-farm incomes by 25%. Tech as a tool for social mobility, in the place I come from.
4 USPTO patents filed
Business Process AI, Regulatory Controls AI, Auto-Remediate AI, and LLM-Native Process Mining engines — four novel systems patented across 2022–2026. I'm here to do original work, not derivative work, and I'm not shy about putting my name on it.
GMAT 770 (99th percentile)
Maximum percentiles across Quant, Verbal, and Data Interpretation. Flagged not for the score itself but as a signal — rigor is a habit, and the habit shows up everywhere else.