Sandeep Baligeri · AI Architect
Taking AI agents to production — the harnesses, evals and QA that keep them honest.
AI Architect. Builder of Releaslyy. Sourced, specific posts, reviewed by a human.
Latest
All postsAn agent harness checklist: tools, memory, permissions, approval gates
What has to exist around the model before an agent is safe on real work: tools, skills, memory, permissions, budgets, approval gates and evals.
Oct 28, 2026Your CLAUDE.md is probably costing you more than your prompts
My CLAUDE.md grew to 46KB of project history that Claude Code re-sent on every prompt. Here is what it cost me and how I cut it to under 1,000 words.
Jul 19, 2026Why I built Releaslyy — the problem every sprint team ignores
After five years as a tech lead I kept writing the same release update by hand. Here is why I built a tool that writes it for QA, product and customers.
Apr 2, 2026
Things I've built
Releaslyy
Live · FounderAI release notes from GitHub, Jira and DevRev, written for each audience.
releaslyy.com
AI Lens
Architect · QualityKioskAI QA agent that maps a live app into a knowledge graph and designs tests.
ai-lens-nine.vercel.app
Litmus
Architect · QK AI LabsEvaluation, observability and release gates for production AI agents.
unitevals.comAbout
I design the parts around the model that decide whether an agent survives real work: tools, skills, memory, permissions and approval gates, plus the evals that gate each release. Before that I led AI capabilities on Katalon and built products end to end at start-ups.