The Author
Background
Carlos Chinchilla Corbacho. PhD candidate at the University of Salamanca, researching the testing of LLM-based and multi-agent systems. ML engineer in industry. Trust, but Trace covers the engineering side of that work — what survives between research papers and production systems.
This site does not cover, describe, or comment on confidential or proprietary systems or work tied to the author’s employment or institutional affiliations. All content is independent.
- Email hello@cchinchilla.dev
- LinkedIn /in/cchinchilla
- GitHub @cchinchilla-dev
Projects
Open-source tools
For evaluating and orchestrating ML systems — and other ML work.
- 2026In developmentagentanvilContract-driven testing framework for LLM-based and multi-agent systems. Portable across runtimes via a stable contracts package, with AgentLoom as the recommended optional backend.PythonA2APytest
- 2026In developmentagentloomA deterministic LLM workflow orchestrator for production. You define the execution graph; the LLM only fills in content. Circuit breakers, rate limiters, token budgets and OpenTelemetry traces as first-class primitives.PythonPydanticOpenTelemetry
- 2024Archivedvocal-biomarkersReproducible ML pipeline for early COVID-19 and respiratory-disease detection from voice — combines acoustic and biomechanical vocal features. Companion code to the IEEE Access 2024 paper.Pythonscikit-learnlibrosaPraat
Open source
Contributions
Patches, primitives, and bug reports across the agent‑tooling ecosystem.
Press
Selected publications
For academic venues.
- 2025PaperAdvanced Machine Learning and Deep Learning Approaches for Estimating the Remaining Life of EV Batteries — A ReviewBatteries
- 2024PaperApplication of Machine Learning Techniques for the Characterization and Early Diagnosis of Respiratory Diseases such as COVID-19IEEE Access
- 2024PaperAutomated Identification of Cylindrical Cells for Enhanced State of Health Assessment in Lithium-Ion Battery ReuseBatteries