Mohith Korangi Work with me

Mohith Korangi

I build video editors, design systems and AI products, end to end.

Software Engineer (Full Stack) at Trupeer.ai. Four years at Mastercard before that.

Program Jun 2018

2018now

2018 to now, in one reel

Press play, or pick any clip below

Study, four years at Mastercard, a solo run at Trupeer and the products I build on my own time.

Selected work

Products I designed, built and shipped myself, each written up with its architecture, the decisions behind it and what I measured.

AskMyChats saved inbox in dark mode, showing notes, links and forwarded messages from a demo account

Dec 2025 – now

AskMyChats

Save anything from WhatsApp, find it later and ask about it.

to store 40,000 imported messages
1.5 s
added when the same chat is imported twice
0 rows
spec scenarios, each proven by a named test
574
  • React 19
  • TypeScript
  • TanStack Start
  • Postgres
  • Drizzle
  • pg-boss
PanelCut music and timing step: two tracks with beat counts, reading speed and pace controls, and a summary of 60 shots

Oct 2026 – now

PanelCut

Turns a comic PDF into a beat-synced music video, entirely on your computer.

  • Python
  • PyMuPDF
  • OpenCV
ssTitles home screen on a phone: paste your Gemini key, paste your TMDB token, add screenshots

Nov 2026

ssTitles

Turns screenshots of recommendations into a verified watch list.

  • React 19
  • TypeScript
  • PWA

Feb 2026

EndPRs

An agent that turns pull request review comments into commits.

  • TypeScript
  • Claude Agent SDK
  • Next.js

Sep 2026 – now

Local Agentic Harness

The workspace I use to build with AI agents without lowering the bar.

  • Claude Code
  • Agent skills
  • Subagents
6 See every project Including TestYourTrade, each with its architecture and measured results.

What I build

New products, frontend rebuilds and AI features, for teams who care how their product feels to use.

Work with me
  • Frontends from your designs

    Figma to production React and TypeScript, pixel-accurate, accessible and fast. I've built this into an agent pipeline that hits 98–99% visual accuracy.

  • Design systems

    Tokens, a component library and the rules, tests and CI checks that keep every screen on-system.

  • AI features that hold up

    LLM features with verification, evals, usage limits and kill switches, so they stay accurate and affordable in production.

  • Full products, end to end

    From the data model to the deploy: auth, background jobs, webhooks, testing and monitoring, built and shipped by one person.

Writing

Short, practical explainers on JavaScript, React and developer tools, viewed 70K+ times on Medium. These five are the most viewed.

  1. The Best Readme Generators for Your GitHub Profile21K views
  2. How to Handle Unsuccessful Fetch API Calls in JavaScript11.2K views
  3. I Hacked the Chrome Dinosaur Game. Here's How I Did It.2.8K views
  4. Data Structures: MULTI_SET vs SET vs UNORDERED_SET2.2K views
  5. All You Need to Know About the Fetch API in JavaScript1.7K views

All articles on Medium