AI and LLMs: projects, experience and articles
6 projects · 1 job · 9 articles
Case studies
- ImpressãoMais3DOn-demand 3D printing e-commerce20 bounded contexts
- Digital Tech MSOnline print shop with artwork approval and production20 bounded contexts
- DevAgent — the Autonomous KanbanAn agent that reads the card and opens the pull requestCard-to-PR pipeline
- Eurologado — EU Cosmetics Compliance SaaS100+ customers, AI-assisted regulatory dossiers100+ customers
Other projects
- 99Freelas Proposal AssistantChrome extension that writes proposals with AIChrome Web Store
- Ikigai AILLM-guided self-knowledge engine
Professional experience
Senior Software Engineer & AI Developer · Independent
Jan 2026 - present
Strategic work developing autonomous products, Generative AI-powered tools, and advanced corporate workflow automation. Architect and creator of DevAgent (autonomous software engineering system).
Articles
- AI Shouldn't Handle the Money: Modeling RetomaVenda with Ledgers, Leases, and IdempotencyHow to surround a probabilistic component with deterministic invariants for credits, privacy, idempotency, leases, and cost in an AI product.
- Does Clean Architecture still matter when AI writes the code?Coding agents made code cheaper to produce. That did not make architecture, readability, and testing less important — it changed where those principles need to be enforced.
- Kimi K3, one million tokens, and the context trap: more memory is not more architectureKimi K3 and K2.8 push coding agents toward huge context windows. Useful, yes—but loading an entire repository is not a substitute for retrieval, tools, tests, and a good harness.
- In 2026, the model is only one component: the agent architecture I would actually run in productionKimi, Jev, the Agents API, and modern coding agents make one thing clear: differentiation is moving from the standalone model to the harness, tools, controls, and verification.
- Jev, decisions instead of strings: why not every agent step needs an LLMJev proposes a different layer for agents: typed decisions, probabilities, and confidence. The point is not replacing LLMs, but stopping the habit of using them everywhere.
- MCP in production in 2026: I stopped thinking about tools as functions and started thinking about boundariesWhat changed in how I understand MCP: stateless protocol design, authorization, tools, private environments, Tasks and why connecting an agent to real systems is an architecture and security problem.
- Why I am learning Python for AI without abandoning TypeScriptAfter years of JavaScript and Node.js, I added Python and FastAPI to my AI stack—not because of hype, but because different ecosystems deserve different boundaries.
- From backlog card to pull request: what I learned building a real software agentThe architecture and trade-offs behind DevAgent: an agent that reads a task, changes code, runs checks, and delivers a pull request humans can review.
- AI in compliance without outsourcing the decision: what I learned building a PIF SaaS for European cosmeticsHow I modeled Eurologado: AI for drafting and organizing, structured rules for validation, versioning, evidence and audit trails in a regulatory domain.
Used together with
Need AI and LLMs on your team or project?
I am open to remote roles and custom projects.
How I work