Senior Software Engineer - Syncade Developer

Job description

At Bristol Myers Squibb, our employees often ask, “Who are you working for?”—a question that fuels collaboration, accountability, and urgency in our work. Our purpose-driven culture inspires us to discover, develop, and deliver innovative medicines to prevail over serious diseases. We offer uniquely interesting and meaningful work, opportunities for growth, and a supportive environment that values inclusion, wellbeing, flexibility, and comprehensive benefits. This is work that transforms the lives of patients, and the careers of those who do it. Position: Sr. Software Engineer – AI Engineer At Bristol Myers Squibb, we are inspired by a single vision – transforming patients’ lives through science. Each of our colleagues contribute to innovations that drive meaningful change. Scope The EMES core platform, its components, interfaces, data layer and tools – Material and Inventory, Order Management, Equipment and Instrument Integration, Sample Management, Patient Scheduling System integration, Label Printing, Recipe Authoring and Execution, Electronic Batch Records, and Batch Reporting and Analytics – together with: AI and GenAI services embedded in manufacturing workflows: assistants, retrieval over controlled documents, anomaly detection and decision support. Web applications and operator-facing interfaces across shop-floor, quality and supply-chain personas. Enterprise data models, databases, audit-trail architecture and reporting data flows; cloud migration and adoption of next-generation manufacturing systems. Key Responsibilities AI Engineering Design, build and operate production-grade AI services – not prototypes – including retrieval-augmented generation over SOPs, batch records, deviations and equipment documentation, and agentic workflows that act against MES systems under human oversight. Own the retrieval architecture (chunking, embedding selection, vector storage such as pgvector or OpenSearch, hybrid search, re-ranking, citation fidelity) and LLM application patterns: structured output and tool calling, MCP servers over enterprise systems, orchestration, cost controls and guardrails. Establish AI evaluation and observability: golden datasets, offline and online evals, regression gates in CI, hallucination and drift detection, tracing and human-in-the-loop review. Apply classical ML where it is the better tool (anomaly detection, forecasting, predictive quality) and ensure responsible, compliant AI: data classification, IP and PII protection, model and prompt versioning, reproducibility and validation evidence for GxP-relevant use. Web Application Development Deliver full-stack applications: React with TypeScript front ends, component libraries and design systems, and responsive, accessible (WCAG) interfaces built for gloved, time-pressured shop-floor use as well as office users. Build backend services in Python (FastAPI) and/or Java (Spring Boot) exposing REST and GraphQL APIs with OpenAPI-first contracts, secured with OAuth 2.0/OIDC, JWT, SSO and fine-grained RBAC, and write the data access layer with the same rigour as the API (deliberate query design, sound ORM boundaries, pagination, no N+1 access, idempotent writes). Implement real-time and AI-assisted experiences – WebSockets/SSE for live batch and equipment status, token streaming with inline citations – alongside validation-rich workflows including e-signature interaction patterns. Own quality and performance end to end: automated tests across the stack (pytest, Jest/Vitest, Playwright), load testing, and CI/CD (GitHub Actions, Jenkins, ArgoCD) with GitOps delivery of application and database changes together against measured p95 targets. System and Data Architecture Lead the design of distributed, multi-tenant systems: bounded contexts and Domain-Driven Design, service decomposition, API and event contracts, and evolution strategy. Own the data model end to end – conceptual, logical and physical design, deliberate normalization, and referential integrity, constraints and domain invariants enforced in the database rather than only in application code. Demonstrate advanced relational expertise: complex SQL, window functions, CTEs, stored procedures (PL/pgSQL, PL/SQL), partitioning, deliberate indexing strategy, and performance diagnosis from execution plans, statistics, lock contention and pool exhaustion. Apply correct transaction design (isolation levels, concurrency control, idempotency, retry semantics) and migration discipline (Flyway/Liquibase/Alembic, expand-contract zero-downtime rollout, tested rollback), and run databases as production assets on AWS: RDS/Aurora PostgreSQL, Multi-AZ, read replicas, pooling, PITR and DR. Design regulated-grade data integrity – immutable, append-only audit trails (including hash-chained designs), e-signature records, ALCOA+, retention and archival – satisfying GxP, 21 CFR Part 11 and GAMP 5 (2nd edition, incl. AI/ML guidance) by design, not documentation alone. Design event-driven systems with Kafka (partitioning and ordering guarantees, transactional outbox, CQRS and event sourcing, sagas) and enable analytics without burdening the transactional system through change data capture (Debezium) and dimensional models for batch reporting. Reason rigorously about non-functional requirements – capacity, scalability, SLOs, failure modes, circuit breakers, backpressure, graceful degradation when a site link is down – and define caching, queuing and throttling strategy (Redis, SQS/SNS, idempotency keys). Architect for hybrid deployment (on-premise and near-line where shop-floor latency demands it, AWS for scale) using Docker, Kubernetes (EKS/ECS) and IaC; instrument for observability (Prometheus, Grafana, OpenTelemetry); and apply security by design, mitigating the OWASP Top 10 and OWASP Top 10 for LLM Applications. Collaboration and Leadership Partner with Manufacturing, MS&T, Quality, Automation and site IT to translate process needs into system and data design; produce and defend the artifacts of real architecture work (ADRs, design documents, data models, sequence and data-flow diagrams, threat models) and run design reviews; and communicate integration contracts to technical and non-technical stakeholders. Mentor engineers, review code and designs, and raise the bar on clean code, TDD/BDD and craftsmanship – including governed AI-assisted development that improves throughput without weakening review rigour or compliance – while driving continuous improvement through Agile/Scrum across global teams. Required Qualifications & Skills Bachelor’s or Master’s degree in Computer Science, Software Engineering or a related field, with 10-12 years of software engineering experience, including 5+ years building and operating web applications and 2+ years delivering AI/ML or GenAI capabilities into production (prototypes alone will not qualify). Expert-level SQL and relational data modeling with deep hands-on PostgreSQL (or Oracle/SQL Server): schema design, indexing, query tuning, transactions and migrations. Demonstrable system design capability at the level of a multi-service, multi-tenant platform – bounded contexts, API and event contracts, consistency and failure modes, capacity and scaling – with Docker, Kubernetes (EKS/ECS), cloud-native architecture on AWS, infrastructure as code and Kafka-based event streaming. Strong hands-on Python 3.x with FastAPI and TypeScript with React; Java/Spring Boot experience is a strong advantage. Practical experience with the LLM application stack: model APIs, embeddings and vector search, RAG, tool calling, prompt and context engineering, and evaluation. Agent frameworks (LangChain/LangGraph) and MCP server development are an advantage. Working knowledge of MES/EMES platforms, pharmaceutical manufacturing and regulatory compliance – ISA-88 and ISA-95, electronic batch records, recipe execution – or demonstrated ability to ramp quickly in a regulated domain. Solid grounding in application and data security (OAuth