In this role, you will:
New ML opportunities get identified and proven out before engineering time is spent on them, because you research, prototype, and validate the model first.
Complex ideas land clearly across teams, because you can explain a model's logic and tradeoffs to engineers, product, and stakeholders without losing the substance.
Models keep working after they ship, because you own the full lifecycle: development, production deployment, drift monitoring, and retraining as the data changes.
What you Bring:
3+ years' experience as a Data Scientist, working with Python, SQL, and the standard data science toolkit (Jupyter Notebook, Pandas, scikit-learn, TensorFlow, PyTorch).
BSc in an exact science: mathematics, computer science, or statistics
Experience building prediction and clustering models using both supervised and unsupervised methods.
Proven ability to own the algorithm/data science lifecycle end to end, from idea to production.
Experience running models in production: feature/prediction drift analysis, alerting, and updating models to work with the latest data
Familiarity with the MLOps lifecycle.
Comfortable using AI-assisted development tools (Claude Code, GitHub Copilot, Cursor) to speed up experimentation and iteration.
Working knowledge of GenAI beyond coding assistants: prompt engineering and building solutions on top of LLMs/multimodal models, since some of our production problems are solved with an LLM rather than a traditional model.
Comfortable working independently on abstract, loosely-defined problems in a fast-moving, agile environment.
Good to have:
Experience with routing and navigation algorithms.
Experience with Vertex AI or an equivalent cloud ML platform (training, deployment, monitoring).
Experience engineering geospatial/location-based features (geohashing, lat/lng, zip-code and polygon-based features, geofencing) for real-world prediction models.
