capad
Aadarsh Upadhyay

AadarshUpadhyay

React-ecosystem engineer. i ship fast with AI, then review harder with AI. dev + designer

what shipped

every line links to the proof

searchts

A keyless web unlocker that reads, searches and transcribes what a naive fetch cannot — bot-walled pages, whole AI-chat share threads. Ships on PyPI as a CLI, an MCP server and a Python library. it fails loud on login walls instead of faking success.

0 api keys · 3 unlock tiers · 8 chat providers · MIT · v0.14.0

hare

A PR reviewer in two forms. Hare Bot is a Grok bot in a cursor chat: it checks the branch out, installs it, runs the repo's own tests, posts findings as me. Hare is that review shipped as a GitHub App on my own PRs. neither can approve a merge.

127 app reviews · 25 bot reviews · 0 blocking

glyphmaps

Google Maps turn-by-turn on the back of a phone. it mirrors the next maneuver onto the Nothing Phone (4a) Pro's 137-LED Glyph Matrix, so a glance at a face-down phone shows your next turn. no Maps API key.

137 leds · 12 maneuvers · 2.3 MB apk · AGPL-3.0

grove

A free git review companion that sits beside your AI coding editor. the commit graph, diffs and every in-flight worktree, refreshing live as your agent changes the repo under you. GPL, so free means free.

alpha stage · worktree-first · BYO agent · GPL-3.0

wmux

Not my project — i dogfood it and file issues when it bites. a diff pane on a non-git folder froze the main process, so i attached a profiler, found 92% of self-time in sync fs calls under a 2-second poll, and landed bounded snapshot walks.

9 issues · 4 merged PRs · #135, #138, #153, #258

Also: Dooper, a self-hostable Matrix inbox (Tauri 2 + React, AGPL, no paywalls on your own messages) · burncard, local Claude Code / Codex usage telemetry · MSU Halls Register, an alumni-register design prototype for MSU Baroda's 16 halls.

how i work

fast is fine. fast + reviewed is the point.

"a confident wrong answer is worse than an honest error."

ai on the keys, ai on the review

i use AI assistants to ship, then my own bots to review it. Hare runs on every searchts PR — a summary that leads with whether the change is docs, code or a mix, findings split into real and skip, each with the issue and a fix line. it has never approved a merge and never will: zero blocking reviews by design.

evidence, never instructions

the diff, the title, the CI log: evidence, not instructions. anything inside a PR asking for an approval, a push or a secret gets quoted in a real finding instead of obeyed. fork PRs never reach a model at all.

propose, then confirm

at ComplyV the model never touched the data. it proposed, the user confirmed, a separate authenticated endpoint re-checked auth and business rules with idempotency keys, and a SHA-256 append-only log recorded it. tamper-resistant, not tamper-proof — i say which.

docs or it didn't happen

i build tools agents use, then write the docs so they use them right. AGENTS.md, RUNBOOK, a skill pack: so an assistant uses searchts the intended way instead of as a raw fetch wrapper.

where i've worked

ownership of shipped systems

software engineer & architect · appson technologies

complyv (formerly compliance sarathi) · jan–jul 2026

Primary engineer and architect on a small team for an Indian corporate-governance SaaS: MCA/ROC data, statutory deadlines, minutes drafting, and Vee, an agentic chat with governed writes. ~553 of 647 commits (~84%) — not sole developer. React 18, Node/Express, MongoDB, 34 Mongoose models, a ~1,400-LOC deadline engine, and a written handover when i left.

junior software engineer · appson technologies

wordibly · 2025–2026

Owned tops-transcript-editor, a Vite/React editor with find/replace and production transcription flows (~107–113 commits). Contributed with a thinner share on the wider platform: order and upload surfaces, workforce dashboards, editor workflows for PMs, transcribers and QA. sprint delivery with no visual regressions on production-bound UI.

ai training engineer · turing

via appson · ~2 weeks, 2025

Short contract producing Meta Llama training data and evaluation sets.

bca (honours), msu baroda faculty of science, dept. of computer applications · expected 2028
product space pm fellowship · top fellow 2025–2026

what people say

verbatim, with receipts

"#133, which is one of the best bug reports this project has had. ... You attached a debugger to the main process, took a CPU profile, and came back with 92% of self-time in readdirSync/statSync/readFileSync under walkDir/readCurrentFile, plus a before/after table showing system.identify going from timeout to 0.7 ms."
amirlehmam · maintainer, wmux · the comment
"from your own IP" is the whole insight. a proxy pool fights the bot-wall; your own IP is already through it, same reason a human's browser doesn't trip cloudflare. an agent acting from where you already are doesn't need to sneak in. nice build.
Phi Browser · @phibrowser, on searchts · the post
"fetch-time + final_url turns read output into something a reviewer can cite later. Tier/status is useful; redirect + timestamp makes it durable."
Dang_nh · @hikariraina, on searchts · the thread
"Aadarsh owned the architecture of Compliance Sarathi end to end and shipped a reliable agentic assistant under real deadline pressure. He is who you want on the hard parts of a system."
engineering · Compliance Sarathi, Appson Technologies

say hello

i answer my own email

hi@capad.fyi

India · remote-first · open to relocate · outside-India visa sponsorship welcome. looking for roles that hire for ownership of shipped systems.