Senior AI Data Scientist

Senior AI Data Scientist (6+ years)

• Own the full ML lifecycle for 12 highimpact problem areas (discovery delivery iteration).
• Translate ambiguous business problems into testable hypotheses and modelable objectives.
• Design, train, evaluate, and ship models (incl. LLM/RAG, forecasting, ranking, NLP/CV as needed).
• Build robust offline/online evaluation (A/B tests, counterfactuals, guardrails, humanintheloop).
• Productionize models with softwareengineering rigor (versioning, CI/CD, monitoring, observability).
• Drive data quality: schema design, feature stores, labeling strategy, data contracts, governance.
• Communicate clearly with execs and nontechnical partners; influence roadmaps with data.

Minimum Qualifications

• 6+ years in applied ML/Data Science with production impact (consumer or enterprise).
• Strong statistical foundations (causal inference, experimental design, uncertainty, metrics).
• Proficiency in Python and scientific stack (pandas, NumPy, scikitlearn); SQL fluency.
• Experience training and evaluating modern ML (treebased methods, GLMs, embeddings, deep learning).
• Experience deploying to production with at least one MLOps stack (e.g., MLflow, Kubeflow, SageMaker, Vertex, Databricks) and containerized services (Docker/Kubernetes).
• Track record of partnering with engineering and product; excellent written and verbal communication.

Preferred (Nice to Have)

• LLMs and Retrieval Augmented Generation (RAG): prompt/retrieval design, evaluation, safety/guardrails.
• Generative AI: finetuning, adapters/LoRA, instruction tuning, synthetic data, vector DBs (FAISS/PGV/Chroma).
• Recsys/ranking, timeseries forecasting, or causal ML for growth/marketing experimentation.
• Data engineering exposure (dbt, Spark, Kafka) and production telemetry/monitoring (Prometheus, Grafana, Evidently).
• Experience with privacy, responsible AI, and model risk management.

Core Tech & Tools (illustrative)

• Languages:

  Python, SQL.

• Libraries:

  scikitlearn, XGBoost/LightGBM, PyTorch/TensorFlow, Hugging Face, Ray.

• Data:

  Snowflake/BigQuery/Redshift, dbt, Airflow.

• MLOps:

  MLflow, Weights & Biases, Docker, Kubernetes, Feast/Featureform.

• Serving:

  FastAPI, gRPC, AWS/GCP/Azure serverless, vector DBs.

• Analytics/Experimentation:

  Amplitude/GA, Optimizely/Statsig, A/B testing frameworks.