Responsibilities

• Design and implement an evaluation strategy for AI agents within the Databricks environment.
• Apply data science methodologies to assess model and agent performance, reliability, and business relevance.
• Contribute to MLOps workflows supporting the development and deployment of AI agents.
• Contribute to AI solution refinement through experimentation, validation, and optimization.
• Prepare and deliver client demos, including presentation preparation, demo planning, and showcasing AI use cases.
• Perform basic domain research (chemistry-related) to support demo preparation and solution demonstrations.

Requirements

• Strong Python engineering skills, including writing production-quality code and tests.
• Experience building and integrating REST APIs (e.g., FastAPI/Flask) for AI services.
• Practical experience with LLM agents / agentic workflows, including state management, tool/function calling, iterative execution, and output validation/guardrails.
• Familiarity with core ML tooling (e.g., NumPy, scikit-learn, PyTorch) sufficient to support experimentation and validation.
• Experience working with data stores (SQL databases such as PostgreSQL/MySQL, object storage such as S3, and/or NoSQL).
• Working knowledge of MLOps / LLMOps: experiment tracking, versioning, CI/CD for AI workflows, and deployment/monitoring basics.
• Experience designing evaluation and QA frameworks for LLM/agent outputs (quality, accuracy, consistency), including offline test sets, metrics, and regression testing.
• Hands-on experience building production LLM applications and agents (RAG, multi-step reasoning, tool/function calling, MCP).
• Experience working with Databricks.
• Experience working with Docker.
• English level B2+.

Nice to have

• Experience with cheminformatics (e.g., molecular fingerprints, chemical descriptors, RDKit, molecular similarity search).