Open to Applied AI · Forward Deployed · Founding · Staff

Sérgio Brito
AI Agent Systems Engineer

I build production multi-agent systems: MCP tools, multi-LLM orchestration, eval frameworks, tier-enforced safety, and autonomous Claude Code workers. 10+ years shipping end-to-end — backend, mobile, infrastructure. Multiple AI-native products running in production today.

AI Agent Systems Multi-Agent Orchestration MCP / Tool Use Multi-LLM (Claude · GPT · Gemini) RAG & Eval Frameworks Claude Code SDK Python Ruby on Rails TypeScript / Node.js Multimodal Pipelines React / React Native PostgreSQL · pgvector AI Agent Systems Multi-Agent Orchestration MCP / Tool Use Multi-LLM (Claude · GPT · Gemini) RAG & Eval Frameworks Claude Code SDK Python Ruby on Rails TypeScript / Node.js Multimodal Pipelines React / React Native PostgreSQL · pgvector
View Projects Get in Touch LinkedIn 📄 CV / Resume
0 Tools in production
0 AI-native systems shipped
0 Years shipping software
0 Records/day at scale
About Me

Production AI
from First Principles

Multi-agent orchestrators, MCP servers, multimodal pipelines, and the boring stuff that keeps them running in production — observability, evals, safety, deploy.

Sérgio Brito

Sérgio Brito

AI Agent Systems Engineer · 10+ Years Production

Senior engineer with 10+ years turning ideas into shipping products. Today I focus on Applied AI: multi-agent systems coordinated by Claude / GPT / Gemini, tool-call loops with MCP, eval frameworks, and tier-enforced safety. Before that, Rails / Node / Python at scale for fintech, healthtech, e-commerce, government — millions of requests, 99.9% uptime, the works.

Currently running an in-house multi-agent orchestrator (3 coordinated agents, 20+ MCP-style tools, autonomous Claude Code workers, audit-grade event log) that runs my own ops. Earlier built and shipped 'Vik Nutri' AI to 1,000+ users at VIK with function calling and per-user memory. Multiple AI-native products live in production. Comfortable in the messy middle: from prompt to pipeline to PagerDuty.

Location Campo Grande, Brazil · Remote (US-friendly TZ)
Education B.Sc. Software Engineering — UFMS
Experience 10+ years · production AI since 2022
Languages Portuguese (Native), English (C1 Advanced), Spanish (Advanced)
Projects

Production AI Systems

Multi-agent orchestrators, multimodal pipelines, AI-native products. Most are private — what's public below is architecture and outcome, not code.

Flagship · Multi-Agent · MCP Production · Private

Bailder

Personal multi-agent orchestrator. WhatsApp-native, autonomous Claude Code workers, tier-enforced safety.

An in-house AI workspace that runs my own ops — 3 coordinated agents, 20+ MCP-style tools, autonomous Claude Code headless workers, tier-based safety enforced in code (green / yellow / red), audit-grade event log, multi-LLM routing via OpenRouter, Playwright MCP for visual e2e. Bootstraps brand-new projects (repo + DB + DNS + deploy) from a single tool call. Private repo by design.

Ruby on Rails 8 Claude Code SDK MCP (Playwright) Multi-LLM (OpenRouter) Solid Queue WhatsApp Cloud API DigitalOcean (auto-deploy)
  • 3 coordinated agents · 20+ tools · autonomous Claude Code workers
  • Tier-enforced safety in code (PreToolUse hooks), not in the prompt
  • End-to-end project bootstrap from a single tool call (repo → DB → DNS → deploy)
  • Event-sourced audit log · scheduled smoke evals · circuit breakers
Architecture
Multimodal · Sports Live

SelfScore

Amateur football analytics. Multi-camera video pipeline + GPT Vision + real-time WebSocket coaching.

Multi-camera synchronization plus a 5-phase video pipeline (FFmpeg → chunking → Whisper STT → Gemini Vision segmentation → AI analysis) ingests footage, extracts plays, and powers real-time analysis with 8 simultaneous tool calls over WebSocket. 5 AI narrator personalities with per-player profiling.

Rails 8 + React Native (Expo) Gemini Vision FFmpeg Whisper WebSocket (8 tools in-flight) Sidekiq
  • 5-phase video pipeline: chunking → STT → segmentation → cutting → AI analysis
  • Live analysis with 8 simultaneous tool calls via WebSocket
  • Per-player coaching profile · 5 narrator personalities
Live →
Multi-Agent · Sales Production

Giants

Multi-agent sales platform. Multi-channel agents, human takeover, anti-promise detector.

Sales & support widget powered by a multi-agent stack across WhatsApp, Instagram, and Facebook. Agents on Claude / GPT / Gemini with structured handoffs to humans and an anti-promise detector that flags risky outputs before they ship to a lead.

Rails Multi-LLM (Claude · GPT · Gemini) WhatsApp · Instagram · Facebook Hotwire PostgreSQL
  • Multi-LLM agents (Claude / GPT / Gemini) with channel-aware routing
  • Human takeover protocol with full conversation context
  • Anti-promise detector pre-publishes risky outputs
Infra · LLM Gateway Production

Railter

Internal LLM gateway. OpenAI-compatible, SSE streaming, per-key usage & billing.

Central API gateway for the whole ecosystem. Unifies Claude, GPT, Gemini behind one OpenAI-compatible interface with SSE streaming, per-key auth, usage dashboards, and billing. Every other product calls Railter instead of the LLM providers directly.

Rails 8 OpenAI-compatible API SSE Streaming Multi-LLM Routing Usage Metering
  • Unified OpenAI-compatible API for Claude / GPT / Gemini
  • Per-key auth, usage dashboard, billing
  • SSE streaming with retry / fallback
Multimodal · Personal AI Production

Chailter

Personal multimodal AI workspace. Shared agents, scheduled workflows, Kids mode.

ChatGPT-like personal workspace with multimodal input (audio, image, PDF), pre-configured agents, recurring workflows (cron / webhook / email triggers), and a safety-tuned Kids mode. Consumes LLM via Railter. Solid Queue + Hotwire + Tailwind v4.

Rails 8 Hotwire · Tailwind v4 Solid Queue Whisper · Gemini Vision Railter (LLM gateway)
  • Multimodal input pipeline (audio · image · PDF)
  • Recurring agents on cron / webhook / email triggers
  • Kids mode with safety-tuned guardrails
Conversational · Dev Studio Production

AircTech

AI-powered dev studio. Conversational scoping agent with function calling.

AI-powered development studio. Automated project scoping via a conversational agent with function calling and structured outputs. The agent drives discovery, scores project viability, and manages the full client lifecycle.

Rails 8 React Native (Expo) OpenAI Claude PostgreSQL
  • Conversational discovery agent with structured requirement extraction
  • Viability scoring & client lifecycle management
Marketplace · Mobile Live

Free Diária

Gig-work marketplace. Intelligent matching, swipe discovery, PIX wallet.

Solo-built gig-work marketplace with intelligent professional matching, swipe-based discovery UX, digital wallet with PIX integration, bidirectional reviews, and real-time chat.

Rails React Native Stripe + PIX Google Maps PostgreSQL
  • Intelligent matching · swipe discovery · bidirectional reviews
  • Digital wallet with PIX (Brazilian instant payments) integration
  • Real-time chat between gigs
And more... Production

Other AI Systems

Organaizer, BBE RAG, Mailder, Hetal Retail, and a few dozen production agents.

Organaizer (multimodal personal assistant with Whisper + Vision + tool calling), BBE Tour Generation (RAG over proprietary tour catalogs), Mailder (multi-tenant email SaaS with SPF/DKIM/DMARC), Hetal Retail (120K+ test lines), and dozens of production agents shipped for real businesses.

Whisper GPT / Gemini Vision pgvector / RAG Tool Use Multi-Tenant SaaS
  • Organaizer: multimodal personal assistant, DB-driven capabilities, tool calling
  • BBE: RAG over proprietary catalogs with pgvector
  • Dozens of production AI agents shipped for real businesses
Explore all projects →
Skills

AI-First Tech Stack

Production tooling for multi-agent systems, plus the full-stack engineering chops to ship them end-to-end.

🤖

AI Agent Systems

Multi-agent orchestration in production. Tool-call loops, sub-agents, autonomous Claude Code workers, MCP servers, context engineering.

Multi-Agent Coordination MCP (Model Context Protocol) Function Calling / Tool Use Claude Code SDK Sub-Agents & Hand-offs Context Window Management
🏗

Multi-LLM Orchestration

Routing, fallback, and cost-aware orchestration across providers. Streaming, structured outputs, retries, circuit breakers.

Claude (Sonnet · Opus) GPT-4o / GPT-5 Gemini Flash / Pro Llama · Mistral OpenRouter / LiteLLM SSE Streaming
🎬

Multimodal Pipelines

Video, audio, image with AI — from capture to insight, in batch or real-time. WebSocket streaming, 8+ tools in flight.

Video Processing (FFmpeg) Computer Vision (GPT / Gemini Vision) Speech-to-Text (Whisper) Image Generation (Gemini Flash) Real-Time Streaming WebSocket / SSE
✅

AI Safety & Evals

Safety enforced in code, not in the system prompt. Tier-based guards, audit logs, eval gates, circuit breakers, fallbacks.

Tier-Based Tool Guards Eval Frameworks (custom + RAGAS) Audit & Event Sourcing PendingConfirmation Flows Pre-Tool Hooks Smoke Evals on Schedule
⚙

Backend & APIs

Robust APIs and scalable systems with a focus on correctness, performance, and operational sanity.

Ruby on Rails 8 Python (FastAPI) Node.js / TypeScript REST · GraphQL · SSE Sidekiq · Solid Queue Background Jobs
📱

Frontend & Mobile

Hotwire-first for AI-native dashboards, React Native for mobile. Real-time UIs that don't fight you.

Hotwire / Stimulus / Turbo React / TypeScript React Native / Expo Tailwind CSS WebSocket UIs
📊

Data & Vector Stores

Schema design, query tuning, RAG over proprietary data. pgvector-first, lifted from production scars.

PostgreSQL (tuning · partitioning) pgvector / RAG BigQuery · Looker Elasticsearch MongoDB · MySQL ETL & Data Migration
☁

Cloud & DevOps

Opinionated infra. Auto-deploy from git, observability that survives outages, cost-aware multi-tenant patterns.

DigitalOcean (App Platform · DBs) AWS (S3, RDS, ECS, Lambda) Docker · Compose Kubernetes GitHub Actions · GitLab CI Systemd Auto-Deploy
🏗

Distributed Systems

Event-driven architectures, bounded contexts, async-first messaging. Scaled monoliths into services the right way.

Event-Driven Architecture Kafka · RabbitMQ Microservices · DDD CQRS / Event Sourcing System Design
🧭

Engineering Craft

Patterns, models, and clean architecture — the stuff that makes the codebase outlive the deadline.

Design Patterns · SOLID Clean Architecture Data Modeling TDD · BDD (RSpec · Jest) Refactoring at Scale
👥

Leadership & Delivery

Tech leadership, mentorship, code review, and async-first delivery — built AI-augmented dev culture at scale.

AI-Augmented Dev (Cursor · Claude Code) Mentorship & Code Review Technical Leadership Shape Up · Agile · Scrum Async-First Collaboration
Experience

Professional Journey

Over a decade building production systems — increasingly AI-native since 2022.

Aug 2025 – May 2026

Senior Data Migration Engineer · AI-Augmented

Monument (Remote, US)

Self-storage SaaS startup. Led migration for dozens of enterprise clients, hundreds of facilities, thousands of units — using Claude Code as primary dev driver.

  • Claude Code as primary dev driver: agentic scaffolding, refactoring, migration diff review
  • Idempotent pipelines with multi-level validation reconciled to the cent before cutover
  • Internal tooling that significantly accelerated migration speed across clients
Claude Code TypeScript Node.js MySQL AWS
May 2023 – Jul 2025

Principal Software Engineer · AI Platform Lead

VIK (Remote)

Enterprise B2B healthtech. Designed and shipped the company's AI platform from zero; led platform re-architecture and distributed systems.

  • Shipped 'Vik Nutri' AI to 1,000+ users — function calling, per-user memory, multi-LLM routing
  • Designed company AI platform from zero: agents, RAG, multi-LLM orchestration via OpenRouter / LiteLLM
  • Drove company-wide AI-augmented dev adoption (Cursor + Claude Code conventions)
  • 99%+ performance gain on enterprise dashboards via query tuning, async jobs, Redis caching
  • Monolith → distributed: bounded contexts, microservices, event-driven messaging (Kafka, RabbitMQ)
Ruby on Rails React OpenAI · Claude · Gemini OpenRouter · LiteLLM Kafka · RabbitMQ BigQuery · Looker
Nov 2022 – May 2023

Senior Software Engineer (Consultant)

META | YOUSE — Caixa Digital Insurance (Remote)

Brazil's largest digital insurer. Microservices inside a 50+ service event-driven architecture; critical legacy data migration for risk assessment.

  • Microservices integrated into 50+ service Kafka-based event mesh
  • Legacy data migration (CSV/JSON to unified schemas) critical for risk assessment
  • React dashboards for claims analytics; AWS deployments with CI/CD
Node.js Ruby React AWS (S3, RDS, ECS) Kafka
Sep 2021 – Nov 2022

Software Engineer

Adminer (Remote, B2C)

E-commerce analytics platform serving B2C at scale — millions of requests/day.

  • 'Adminer Analytics' Chrome extension with 50,000+ downloads
  • Async data processing at scale; proxy orchestration across PG/Mongo/ES
  • Frontend revamp (Angular + Tailwind); CI/CD reducing deploy errors by 90%
Node.js Ruby Angular PostgreSQL MongoDB Elasticsearch
Nov 2020 – Nov 2021

Java Software Developer

MJV Technology & Innovation | Bradesco Insurance (Remote)

Critical async batch processing for one of Brazil's largest banks.

  • Async batch processing up to 1M records/day; ETL with Airflow
  • PostgreSQL optimization (indexing, partitioning, query tuning) — cut query time by 60%
Java Spring Boot PostgreSQL Apache Airflow
Jan 2017 – Nov 2020

Java Developer

PSG | DETRAN — Government Traffic Dept. (Remote)

Owned full lifecycle of a government infractions system handling millions of vehicles and integrations with multiple public APIs.

  • High-availability infractions system at multi-million-vehicle scale
  • Internal tooling, relational data models, DevOps on Heroku — 99.9% uptime
Java Spring Boot PostgreSQL Heroku
Contact

Let's Build

Interested in working together? Drop a line — I respond within 24 hours.

Available for AI agent engineering, Applied AI, and senior full-stack roles. Remote · US-friendly TZ.

in
LinkedIn
linkedin.com/in/sergio-brito145
 </>
GitHub
github.com/SergioBrito
📍
Location
Campo Grande, Brazil · Remote

Send a Message

Looking for an Agent Systems Engineer?

10+ years shipping production systems. Today: multi-agent orchestrators, MCP tools, autonomous code workers, eval-gated AI. Open to Applied AI, Forward Deployed, Founding, and Staff AI roles.

Connect on LinkedIn Get in Touch
SB
Sérgio Brito
Online
👤 About 📚 Projects 💻 GitHub ✉ Get in Touch