Agent-native GTMProduct + system designPre-revenue

From a company domain to an outbound pipeline an agent runs

Dark Funnel, agent-native GTM infrastructure

Public-data productEvidence pipelineSide project

Type a company. Get profitability, burn, and headcount from public sources.

Bixbite: sourced financial and hiring signals

System designProduct UXRealtime architecture

A hiring product where the job board updates itself

Fly AI, hiring for small teams

I build agent infrastructure and AI-native products.

Right now, Dark Funnel.

Move your cursor to see it the way an agent does.Keep scrolling. It turns into what an agent sees.

The moat is disappearing

I think the next era of software is built for agents, not humans. Polish commoditizes and taste (what to build and what to cut) doesn't.

Open-source agents are getting very good, and open-source versions of tools like Adobe’s are appearing fast.

Made for agents, used by people

Skills:

  • Next.js
  • TypeScript
  • ORPC
  • LangGraph
  • LangChain
  • Cloudflare
  • AWS
  • Docker
  • Pinecone
  • LlamaIndex
  • GSAP

Outside work, I’m:

  • posting on X, consistently
  • playing chess with friends
  • listening to Dawn Hour by ardley

Straight answers.

What founders usually ask before we talk.

Opal. An AI CRM for job hunting that matches resumes to relevant jobs. I built it and still maintain it.

Real usage: 35+ users, 10,000+ jobs surfaced, 44 resumes parsed, 3,000+ matches. I owned all of it, from the pipeline to the interface.

Try it at opalhq.fun.

Dark Funnel. Agent-native infrastructure for GTM and prospecting. Tell your agent who you want and why now; it finds ICP-fit prospects, verifies the emails, and runs outbound to get your potential buyers.

It builds your ICP from your website, follows up and books the demo, re-checks your list every 3 days, and sends your morning briefing to Slack.

Try it at darkfunnel.pro.

Yes. I've been shipping solo since 2023. I own the path from ambiguous problem to production: orchestration, infrastructure, backend, and UI. I keep the architecture as simple as the system allows and spend complexity only where reliability or performance needs it.

I run a fleet of them in parallel for implementation, debugging, and tests. I set the architecture, review the failures, and decide what ships. Agents increase throughput. Ownership stays with me.

I build around the model: tools, context, routing, memory, evals, recovery, and execution. The goal is reliable behavior in production, including when tools fail.

Full-time founding engineer or AI SWE roles, building agent systems and AI products with ownership from architecture through production. I'm in India (IST).

Email me at [email protected].