Bainbridge - Machine Learning Engineer - LLM/RAG

Role Overview :

The ML Lead owns end-to-end technical direction for machine learning across the platform. Define the model architecture roadmap, establish engineering standards, and act as the bridge between business objectives (recommendation lift, risk reduction, personalization) and production-grade ML systems.

Key Responsibilities :

- Define and drive the ML architecture strategy - from feature store design to model serving patterns - ensuring systems are scalable, maintainable, and production-ready on AWS.

- Lead technical design reviews for recommendation engines, risk scoring models, and personalization pipelines; approve model architectures before development begins.

- Mentor ML Engineers and the MLOps Engineer; set code quality standards, peer-review critical PRs, and hold weekly tech syncs.

- Partner with Head of Data Science on model quality KPIs, experiment design, and A/B test measurement strategy.

- Own ML platform decisions : framework selection (PyTorch vs TensorFlow vs JAX), SageMaker pipeline configurations, feature store tooling (Feast / Tecton), and serving infrastructure (SageMaker Endpoints, Triton, Ray Serve).

- Evaluate and champion adoption of new techniques (transformers for tabular data, causal ML, LLM-augmented scoring) and translate research into production feasibility.

- Collaborate with LLM/GenAI Specialist on integrating generative components into core pipelines - retrieval-augmented generation, embedding-based retrieval, LLM re-rankers.

- Drive incident post-mortems for model degradation events and own the remediation roadmap.

Required Skills & Qualifications :

- 5+ years in applied ML with at least 3 years in a tech lead or staff-level role.

- Deep expertise in supervised, unsupervised, and ranking/retrieval models; solid grounding in probability, statistics, and optimization.

- Hands-on experience with AWS ML stack : SageMaker (Pipelines, Feature Store, Model Monitor), S3, Glue, and Lambda.

- Proficiency in Python (NumPy, Pandas, scikit-learn, PyTorch or TensorFlow); ability to write production-quality, testable ML code.

- Experience building and deploying recommendation or ranking systems at scale (collaborative filtering, two-tower models, LightGBM/XGBoost rankers).

- Familiarity with MLflow or SageMaker Experiments for experiment tracking, and Weights & Biases or similar for model observability.

- Strong understanding of data contracts, feature pipelines (Spark / Flink / Kinesis), and low-latency feature serving.

- Excellent communication skills able to present technical trade-offs to non-technical stakeholders and write clear design documents.

Nice to Have :

- Experience in fintech, credit risk, or financial services ML.

- Familiarity with LLMs (fine-tuning, RAG, prompt engineering) and vector databases (Pinecone, Weaviate, pgvector).

- Knowledge of responsible AI practices : bias detection, fairness metrics, explainability frameworks (SHAP, LIME).

Tech Stack :

- Python

- PyTorch / TF

- SageMaker

- Spark

- MLflow

- AIRFLOW

- CI/CD