Location: Brazil (UTC-3)
Remote: Yes (full overlap with US Eastern, good overlap with Pacific)
Willing to relocate: Within Brazil (São Paulo / Belo Horizonte)
Technologies: TypeScript, React, Next.js, Node.js, Python, PostgreSQL, Supabase, REST APIs, OpenAI API, Claude Code, Vitest, Playwright
Projects: <a href="https://datapsi.com.br" rel="nofollow">https://datapsi.com.br (code: <a href="https://github.com/anthonyhrcr/psych-schedule" rel="nofollow">https://github.com/anthonyhrcr/psych-schedule)
Résumé/CV: <a href="https://linkedin.com/in/anthonyhrcr" rel="nofollow">https://linkedin.com/in/anthonyhrcr (full CV on request)
Email: anthonyhrcr@gmail.com
Full-stack engineer, 3 years in production, always as the only engineer. Looking to join a team where I can grow.
- Municipal education department: I own an enrolment system end to end (React/TS, Node, Postgres). I added an OpenAI classifier with a human review gate. When it started misclassifying, I cleaned the inputs and narrowed what the model could do, instead of prompting harder.
- Regional hospital: I built the integration layer between the hospital information system and clinical devices (monitors, infusion pumps, lab analyzers). Real-time logging, monitoring, and root-causing failures that happened between systems rather than inside one.
- Agenda Psi (<a href="https://datapsi.com.br" rel="nofollow">https://datapsi.com.br): a scheduling and clinical-notes app a psychologist uses every day. React/TS/Supabase. Notes are encrypted in the browser; the master key is wrapped by both a password and a recovery key, so the server only ever stores ciphertext, and row-level security enforces per-user isolation in Postgres. 93 Vitest unit tests and 30 Playwright end-to-end tests run in CI on every PR — writing them exposed a real lost-write bug (an encrypted write that wasn't awaited before the session key was dropped). Code: <a href="https://github.com/anthonyhrcr/psych-schedule" rel="nofollow">https://github.com/anthonyhrcr/psych-schedule
I use Claude Code every day and review everything it writes — a green test suite is evidence, not proof. What I'm missing is code review from people better than me, which is why I want a team.
Open to: full-time, employee or contractor. Junior-to-mid roles. Especially interested in healthcare, gov/civic tech, fintech and AI-product teams.
anthonyhrcr · · focus · HN ↗
Full-stack engineer, 3 years in production, always as the only engineer. Looking to join a team where I can grow.
- Municipal education department: I own an enrolment system end to end (React/TS, Node, Postgres). I added an OpenAI classifier with a human review gate. When it started misclassifying, I cleaned the inputs and narrowed what the model could do, instead of prompting harder.
- Regional hospital: I built the integration layer between the hospital information system and clinical devices (monitors, infusion pumps, lab analyzers). Real-time logging, monitoring, and root-causing failures that happened between systems rather than inside one.
- Agenda Psi (<a href="https://datapsi.com.br" rel="nofollow">https://datapsi.com.br): a scheduling and clinical-notes app a psychologist uses every day. React/TS/Supabase. Notes are encrypted in the browser; the master key is wrapped by both a password and a recovery key, so the server only ever stores ciphertext, and row-level security enforces per-user isolation in Postgres. 93 Vitest unit tests and 30 Playwright end-to-end tests run in CI on every PR — writing them exposed a real lost-write bug (an encrypted write that wasn't awaited before the session key was dropped). Code: <a href="https://github.com/anthonyhrcr/psych-schedule" rel="nofollow">https://github.com/anthonyhrcr/psych-schedule
I use Claude Code every day and review everything it writes — a green test suite is evidence, not proof. What I'm missing is code review from people better than me, which is why I want a team.
Open to: full-time, employee or contractor. Junior-to-mid roles. Especially interested in healthcare, gov/civic tech, fintech and AI-product teams.