My journey in AI started with wanting to break into emerging AI companies, but without deep hands-on experience, it was hard to even get noticed.

I started with scope focused minor projects but then curiosity got expanded to understand what happens under the hood of LLM calls, what goes in & how reliable AI agents works and gets built by world's top companies and how future of agents would evolve in next 10 years when agents will have more controls.

For the last 6+ months, I’ve spent every minute of weekends, nights researching, learning deeper with coding agents and newer models, while managing my day job and family, gained 5 KGs weight along the way.

I wanted to find my Ikigai of what I am good at, what I can sustain working on without loosing my head and effectively also use my 15 years of real work experience from BFSI where serious SOC 2 auditors ask for tamper evident evidences and not just logfile as proof.

That became Tracelane.

With Tracelane, you can host your own neutral model agnostic gateway to put controls, add guardrails, observe everything happening in production environment and proof what your agents did and why.

Though I built everything as solo, I have put every due diligences to ensure the product is secure, fast, ready to use in actual live env by serious adopters.

I would genuinely appreciate your feedback —especially the audit ledger design and what evidence you’d want from an agent recorder before trusting it.