José Paulo Urives Rosa

Senior Software Engineer

Actively looking · Member since 28 Sep 2026

Message

Location: Bauru, Brazil

Desired salary: Confidential

Work preference: Remote Only / Full Time, Part Time, Contract, Freelance

Experience level: Senior

About

Professional summary

I am a senior software engineer with nine years of experience building web, mobile, and SaaS products for Brazilian and US companies. I specialize in designing reliable full-stack systems and have worked in remote-first, US-time-zone-aligned engineering environments.

My strongest expertise is in the PHP and Laravel ecosystem, including Laravel Nova, Filament, and Inertia.js. I also build modern frontend and mobile applications with React, React Native, Next.js, TypeScript, and Node.js.

I have experience delivering large-scale products, including embedded contractor work with Pinterest engineering teams. My work has covered internal tooling, consumer-facing systems, CI/CD pipelines, data workflows, integrations, automated testing, and operational platforms.

I am currently building AI-powered SaaS products from the ground up. My hands-on AI engineering experience includes multi-agent LLM architectures, prompt engineering, cost optimization, model routing, adversarial evaluations, and production safeguards against unreliable model responses.

I enjoy owning systems end to end, from product architecture and database design through deployment, observability, integrations, and unit economics. I am fluent in English and comfortable collaborating closely with distributed teams.

Notice period: Immediately

Tech stack & tools

Working toolkit

Experience

Career history

Founder & Lead Engineer Tutelaris & Parssa

2026 – Present

I am building Tutelaris, an AI-powered jurisprudential research platform for Brazilian lawyers, and Parssa, a WhatsApp-native AI training and nutrition coach. I own the full-stack architecture for these products using Laravel, Inertia.js, PostgreSQL, Redis, WebSockets, React Native, and third-party payment and messaging integrations.

I designed multi-agent LLM workflows, domain-restricted official-source search, model routing by subscription tier, prompt caching, and per-query cost modeling. I also built a deterministic verification layer that compares model responses with database writes, supported by a 60-scenario adversarial evaluation suite used as a pre-deployment gate.

Software Engineer Pinterest (via BairesDev)

2022 – Present

I worked remotely as a full-time embedded contractor with Pinterest engineering teams on large-scale internal and consumer-facing systems serving hundreds of millions of users. My full-stack work used Python, Java, React, Node.js, and SQL.

I built and maintained CI/CD pipelines with Jenkins, Spinnaker, and Buildkite; supported data workflows with Apache Airflow and Apache Hive; and developed internal tools using Retool and GraphQL APIs. I also worked with Cypress, Pytest, and Jest for automated testing, and contributed Next.js development for FreshTracksCanada during an earlier BairesDev engagement.