Skip to content
View berTrindade's full-sized avatar
🏠
Working from home
🏠
Working from home
45 followers · 209 following

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx3Achievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx3Achievement: YOLOAchievement: Quickdraw

Organizations

@poc-micro-frontends

Block or report berTrindade

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
berTrindade/README.md

Hi, I'm Bernardo

I take products from an empty repo to production, and I've been building LLM products since 2024: most recently a voice agent taking live patient calls, built with Python, Twilio and AWS Bedrock.

I work across React and Next.js front ends, TypeScript and Python APIs, and the AWS or Azure infrastructure under them, in Terraform. I like measuring things before claiming them, and writing down why a decision was made.

Latest writing on btrindade.com:

Pinned Loading

  1. llm-engineering-patterns llm-engineering-patterns Public

    Seven runnable LLM proofs of concept with measured results: evals and LLM judges, RAG, tool calling, human-in-the-loop, cost routing, structured outputs, context windows

    Python

  2. twilio-inbound-voice-agent-poc twilio-inbound-voice-agent-poc Public

    Phone voice agent over Twilio ConversationRelay, self-hosted end to end. Latency and accuracy measured: 22 of 24 turns fully correct at a p50 of about 1.8s.

    Python

  3. assistant-channel-poc assistant-channel-poc Public

    An existing product exposed to Claude or ChatGPT as tools. Contended, confirmed and idempotent writes are enforced in code, not left to the model.

    HTML

  4. claude-code-setup claude-code-setup Public

    How I work with Claude Code: commands, sub-agents, skills, hooks and MCP servers, with the workflows behind them.

  5. greenfield-testing-example greenfield-testing-example Public

    Reference CI testing setup for a greenfield service: unit + contract + Testcontainers integration, all in CI

    TypeScript

  6. bullmq-report-generator bullmq-report-generator Public

    Background job processing demo with BullMQ, Redis, PostgreSQL - Async PDF report generation with retries, crash recovery, and real-time status updates

    JavaScript