AI features
AI is a first-class part of how I build products now: Claude Code, Gemini, RAG, reviewers, agents, evals, logs, costs, and rollbacks all in the loop.
I lead product engineering work that has to survive the real world: production AI, performance, reliability, and the judgment calls between idea and shipped system. Outside work it’s music, trips, badminton, gym, and occasionally pretending my Vim config is under control.
Kochi
products / AI
Vim
music / trips / court
places I kept
Yedhin's Favs
$ gh prs --recent
neeto-claude-skills
neeto-cal-web
neeto-bugwatch-web
I’m heading the engineering division at BigBinary, specifically the Neeto ecosystem. My work is less about picking a framework and more about shaping the path: what should exist, what should not, and how to make the final system dependable enough for people to trust it.
These days I’m spearheading the AI push across Neeto: figuring out where AI should sit in the product, building the pieces that make it dependable, and pairing that with performance work and coding-agent workflows that make engineering teams faster without hiding the system underneath.
A few things worth opening next: trip photos, playlist finds, badminton clips, articles, and the work I keep talking about.
TripsMunnar, Kerala

Music16 public picks

Training21 court recordingslatest article
Notes from a year of chasing timeouts on a calendar product that handles millions of records.
recent project index
NeetoChat + NeetoKB AI Assist
<10 secFirst responseNeetoBugWatch
12,000+ReviewsNeetoForm Credential Protection
Auto pauseFake login pagesDec 2024
Dec 2024
Dec 2024
Not a manifesto. Just the tools and stacks that show up in my work a lot: Rails, Postgres, React, AI plumbing, Kubernetes jobs, Vim, tmux, and small scripts that remove boring work.
$ rails console
$ psql
$ tmux
$ vim
$ claude --continue
AI is a first-class part of how I build products now: Claude Code, Gemini, RAG, reviewers, agents, evals, logs, costs, and rollbacks all in the loop.
Rails, React, TypeScript, Sidekiq. The stack I reach for when the thing needs to ship and keep running.
Postgres first. Query plans, indexes, pgvector, and the occasional pganalyze deep dive.
Kubernetes jobs, Docker, tmux, Vim, and enough shell glue to keep boring workflows boring.