Software Engineer - Evals for Agents

Evolve and maintain the agent evaluation platform used by MELI teams. Build offline experimentation tools for real conversation replays and model comparisons. Create a continuous improvement cycle for agents based on evaluation results. Collaborate with product, business, and technology teams to translate needs into impactful solutions.

Strong knowledge of microservices architecture and experience with distributed systems and asynchronous processing. Proficiency in Go and/or Python, with the ability to learn others. Familiarity with relational and non-relational databases, and knowledge of LLMs and their APIs.

Hybrid work modality. Opportunity to work on challenging and innovative projects in a dynamic environment.