About the Role
You’ll be the lead architect of insight, transforming ambiguous business questions into elegant, production‑ready models. Embedded in cross‑functional “mini‑studio” squads with designers and ML engineers, you will shape project vision, own model lifecycle end‑to‑end, and evangelize best‑in‑class data science across client engagements.
Responsibilities
- Design, prototype, and ship supervised, unsupervised, and RL models that deliver measurable ROI
- Build reproducible pipelines (PySpark/DBT/Kubeflow) and champion MLOps standards
- Conduct rigorous experiment design, A/B testing, and causal inference for product decisions
- Translate complex findings into compelling stories for the C‑suite and non‑technical stakeholders
- Mentor junior data scientists through code reviews, reading groups, and paired modeling sessions
- Contribute to internal research agenda; publish white‑papers, speak at conferences
- Partner with solution architects to scavenge and clean multi‑modal data (text, vision, sensor)
- Maintain vigilance for bias, privacy, and responsible AI considerations throughout the workflow
Requirements
- 6 + years building and deploying ML models in production (cloud‑native or on‑prem)
- Fluency in Python, SQL, and one statistical language (R/Stan/Julia); deep grasp of probability theory
- Hands‑on mastery of deep learning frameworks (PyTorch / TensorFlow) and traditional ML stacks
- Proven track record of shipping MLOps: CI/CD, feature stores, monitoring, and drift detection
- Comfortable driving client workshops and presenting to executive audiences
- Experience mentoring teams and setting technical direction
- Bonus: knowledge of GenAI, vector databases, federated learning, or edge deployment