Leticia Mirelly
Lead AI Engineer building production AI systems, multi-LLM workflows, and local-first verification tooling for AI coding agents
Brasília, Brazil
Actions
Leticia Mirelly is a Lead AI Engineer and Forward Deployed Engineer working on production AI systems, multi-LLM workflows, AI cost visibility, and agent reliability.
She has spent the last years building B2B software and AI-native systems across backend architecture, LLM workflows, retrieval, structured outputs, and cost-aware production patterns.
Her current work focuses on the control layer behind AI systems: how teams define scope, measure quality, track cost, understand failures, and decide when AI output is actually ready to trust.
She is currently building Cambone, a local-first AI coding quality gate for teams using tools like Codex, Claude Code, Cursor, and Windsurf. Cambone explores task contracts, attribution, pre-PR checks, repair prompts, and the shift from vibe coding to verified coding.
Leticia writes about production AI, AI coding agents, LLM systems, evals, FinOps, and the engineering decisions that turn AI from a promising demo into reliable work.
Area of Expertise
Topics
From Vibe Coding to Verified Coding: Task Contracts for AI Agents
Coding agents are getting good at producing code. That is no longer the hard part.
The hard part is knowing whether the agent understood the task, stayed inside the right scope, changed the right files, avoided unnecessary complexity, and produced work that can be trusted before it reaches a pull request.
This talk introduces task contracts as a practical control layer for AI coding agents.
A task contract defines what the agent is allowed to do before it starts: intent, allowed scope, forbidden areas, expected evidence, success criteria, required checks, and repair instructions.
I will walk through failure modes I have seen while building Cambone, a local-first AI coding quality gate: no-op changes, scope drift, overengineering, false completion, weak attribution, and agents passing benchmarks while still failing real repo context.
The audience will leave with a concrete task-contract structure they can apply to their own coding-agent workflows.
The core argument is simple: AI coding does not need more vibes. It needs contracts, checks, and control.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top