Key Responsibilities
- Design, build, and deploy ML and GenAI solutions across multiple use cases (e.g., LLM-based search, document summarization, predictive modeling).
- Develop and maintain ML pipelines using MLflow, Databricks, and AWS Sagemaker.
- Implement APIs and microservices for AI model inference using FastAPI and containerization strategies.
- Work with large structured and unstructured datasets (CSV, JSON, PDFs, EMRs, clinical reports).
- Optimize data processing using PySpark, SQL, and RDS (AWS).
- Integrate foundation models and LLMs (e.g., OpenAI, Cohere, Claude, AWS Bedrock) into production workflows.
- Monitor, retrain, and maintain models in production, ensuring high availability, reliability, and performance.
- Collaborate with cross-functional teams (Data Engineering, Cloud, Product) to ensure scalable deployments.
- Document processes, model cards, data lineage, and performance reports for governance and compliance.
Required Skills Qualifications
- 5+ years of hands-on experience in Machine Learning and AI engineering.
- Strong Python programming skills with experience in FastAPI, PySpark, and SQL.
- Hands-on experience with MLflow for experiment tracking, model versioning, and deployment.
- Strong knowledge of Generative AI and LLMs: prompt engineering, fine-tuning, and model integration.
- Experience working on Databricks (Delta Lake, ML runtime) and AWS Cloud services (S3, EC2, Lambda, RDS, Bedrock).
- Familiarity with GitLab CI/CD, containerization (Docker), and cloud DevOps practices.
- Deep understanding of machine learning workflows including data preprocessing, feature engineering, model selection, training, tuning, and deployment.
Preferred / Nice to Have
- Experience in Life Sciences or Healthcare domains: clinical data, RWD/RWE, regulatory AI, etc.
- Familiarity with data privacy regulations (HIPAA, GDPR) or GxP practices in AI workflows.
- Experience working with vector databases (e.g., FAISS, Pinecone) and retrieval-augmented generation (RAG).
- Knowledge of performance monitoring, model drift detection, and responsible AI practices.
