Data Scientist

Must-have: core ML

• Strong SQL: complex joins, CTEs, window functions, and query performance tuning (+4 years experience)
• Strong Python (+4 years experience)
• PySpark or similar (+2 years experience)
• Real forecasting and classification work (demand forecasts, lead scoring, propensity) in production or serious projects. Coursework does not count.
• XGBoost and/or LightGBM in production.
• SHAP to explain and debug models for non-technical stakeholders.

Must-have: GenAI / agentic

• 6+ months of daily, hands-on agentic coding with Claude Code, Codex, or Cursor.
• Has built (not only advised on) GenAI products: RAG, vector search, agents with tool use.
• LLM APIs at scale: batch or async jobs with thousands of calls (data labeling, metadata extraction from text), not one-off prompts.

Must-have: platform

• Azure and/or Databricks (both preferred). Strong GCP/Vertex or AWS/SageMaker experience is acceptable.

Big pluses

• Recommendation engines or personalization.
• Web apps built with coding agents.
• APIs with FastAPI, Docker, and end-to-end deployment.
• React / Next.js.