Job VC

The project is a site-centric Clinical Trial Management System (CTMS) designed to help research sites activate and manage clinical studies efficiently while maintaining full regulatory compliance. The platform serves as a workflow coordination layer that streamlines study start-up, regulatory documentation, sponsor communication, and operational oversight in a validation-ready, audit-compliant environment.
You will play a central role in designing, refactoring, and hardening both the AI and backend architecture to support scalable, compliant clinical operations.
You will collaborate closely with the ML Architect on AI design decisions, work with the Senior ML/Python Engineer on implementation and refactoring, align with the BA on clinical workflow modeling, and support QA in building robust validation and testing strategies suitable for a regulated healthcare environment.
Responsibilities:
•  Design and optimize RAG-based pipelines for medical protocol parsing.
•  Refactor LangChain pipelines for performance, modularity, and cost-efficiency.
•  Implement runtime model orchestration (dynamic models selection based on task complexity and cost — including Vision Language Models, Docling, OpenMed, etc ).
•  Integrate structured validation layers to ensure deterministic transformation from AI output to executable workflows.
•  Enhance vector search performance (through the pgvector embeddings).
•  Implement and optimize various document/content classification ML pipelines
•  Design scalable PostgreSQL schemas for protocol-derived structured data.
•  Model complex clinical entities (Study, Visit, SoA, Eligibility Criteria, Regulatory Tasks, etc.)
•  Implement structured logging, observability, and monitoring of all ML related tasks and pipelines
•  Ensure reproducibility and explainability of AI outputs.
•  Improve automated test coverage (unit, integration, NLP validation tests).
•  Support preparation for computerized system validation (CSV).
•  Enforce secure coding practices and robust secrets management.
•  Maintain containerized deployments (Docker).
•  Improve CI/CD pipelines with structured test stages.
•  Maintain the existing centralized monitoring and error tracking.
•  Ensure system scalability for multi-study, multi-site usage.
Required Skills & Experience
•  5+ years of experience with Python, FastAPI, SQLAlchemy.
•  Strong experience designing microservices architectures and containerized deployments (via Docker).
•  Deep knowledge of PostgreSQL (schema design, indexing, migrations, performance tuning).
•  Hands-on experience with Celery, Redis, and RabbitMQ for async task queues and background job processing.
•  Experience working in healthcare, life sciences, fintech, or other regulated industries.
•  Practical experience with document data extraction
•  Hands-on experience with LangChain and vector databases (pgvector).
•  Understanding of embedding models, AI models orchestration, and cost/performance trade-offs.
•  Familiarity with prompt engineering and deterministic AI output design.
•  Strong system-level thinking and architectural mindset.
•  Ability to balance AI innovation with regulatory reliability.
•  Self-driven, independent, and comfortable in early-stage product environments.
•  Clear communicator in cross-functional teams (product, regulatory, engineering).
Nice to Have
•  Experience designing AI systems in GxP environments
•  Familiarity with pre-commit hooks, Ruff, and Poetry for Python project management.
•  Prior experience in re-architecting systems for high-scale user bases.