HN user

amaze_28

4 karma

building commitposts.com + clubmint.app | backend engineer | shipping in public | anuragsoni.com

Posts0
Comments6
View on HN
No posts found.

the most interesting part here isn't prompt injection worked, it's why the agent had read access to private repo at all while triaging a public issue.

an agent responding to public issue should only ever see context limited to that repo.

it seems like with the evolution of AI - we are slowly missing out basic security practices.

Location: India (UTC+5:30)

Remote: yes — remote-first preferred (worked remotely for 2.5+ years), open to async/global teams, full-time or contract

Willing to relocate: yes (US, EU, UK, Dubai, Singapore — would need visa sponsorship)

Technologies: TypeScript/Node, Next.js, React, React Native, Hono, Cloudflare Workers, AWS (Lambda/Fargate/ECS), PostgreSQL, Redis, LLM systems (OpenAI/Gemini/Groq, RAG, multi-agent)

Résumé/CV: [https://drive.google.com/file/d/1UWk1IXZUFVD59LBV9J2lYDTsUTv...] — work at anuragsoni.com

Email: anuunity28@gmail.com

Backend/full-stack engineer at an edtech serving 15M students. Built the OCR pipeline processing 10M answer sheets bi-monthly (the company's core revenue product), data pipelines at 100–200K records/day, and LLM/RAG systems in production. On the side I build and ship complete products solo — commitposts.com (AI changelogs: GitHub → changelog page, emails, in-app widget) and clubmint.app. Looking for a remote role at a startup where I can own things end to end.