The benchmark resets the repository. Your team does not
Coding agents look more capable when every task starts from a clean checkout. New benchmarks show the cost that appears when patches, decisions, and technical debt carry into the next job.
Developer Experience @ PicPay
Leo Cavalcante
Personal notes on software engineering, developer experience, Go, PHP, infrastructure, and AI.
Recent Posts
AI-generated infrastructure code needs a different review queue
Dockerfiles, pipelines, and Terraform are not just another diff. When AI generates code with access to the production path, review must follow the risk of the file.
GPT-5.6 Luna: the model that became my default in Codex
After nearly 10 billion tokens in Codex, why GPT-5.6 Luna became my default model: cost, performance, context, and limits.
Go 1.27: The Release That Makes Development Feel Smoother
Go 1.27 brings generic methods, smarter tools for testing and maintenance, JSON v2, goroutine leak detection, and a smoother day-to-day development experience.
AI Productivity Is Not One Number: Why the Gains Are Real but Uneven
AI writes code faster. The harder question is whether it became easier to deliver software that people can understand, review, and maintain.
The next leap in AI software development: from assistance to orchestration
The next wave is not just writing code with a copilot. It is coordinating supervised agents, context, tests, and boundaries across the entire development lifecycle.
Ontologies for AI agents: the missing contract between model, memory, and tools
LLMs can converse, but they do not share a stable domain model. Ontologies, knowledge graphs, and semantic validation can give agents a common vocabulary, safer actions, and verifiable memory.
Excessive Agency Is a Reliability Bug, Not Only a Security Risk
When an agent is allowed to act too broadly, the risk is not only leakage or abuse. It also becomes hard to operate: it works outside scope, repeats actions, burns budget, and fails without a clean way back.