Manager AI Engineering

Working with Us
Challenging. Meaningful. Life-changing. Those aren’t words that are usually associated with a job. But working at Bristol Myers Squibb is anything but usual. Here, uniquely interesting work happens every day, in every department. From optimizing a production line to the latest breakthroughs in cell therapy, this is work that transforms the lives of patients, and the careers of those who do it. You’ll get the chance to grow and thrive through opportunities uncommon in scale and scope, alongside high-achieving teams. Take your career farther than you thought possible.

Bristol Myers Squibb recognizes the importance of balance and flexibility in our work environment. We offer a wide variety of competitive benefits, services and programs that provide our employees with the resources to pursue their goals, both at work and in their personal lives. Read more: careers.bms.com/working-with-us.

The GPS Manager will architect and develop enterprise-grade conversational AI systems that enable natural language interactions with analytics, diagnostics, and decision support tools across Global Product Development and Supply. This role focuses on building robust chat orchestration workflows using LangGraph state machines, implementing real-time streaming interfaces, and managing multi-turn conversation persistence at scale.

You will design platforms that combine LLM reasoning with business context, real-time data integration, and conversational UX patterns while supporting autonomous agent workflows. This role requires deep expertise in conversational AI architecture, async Python development, and production LLM integration with sufficient understanding of agentic patterns to enable tool-calling and multi-step reasoning capabilities

Key Responsibilities:

Chat Orchestration & Conversational Workflow Design:
• Design and implement LangGraph-based chat orchestration workflows with typed state management and message reducers
• Build conversational state machines with dynamic routing, context preservation, and graceful error handling
• Develop multi-turn conversation flows integrating LLM reasoning, memory management, and conversation summarization
• Create reusable chat node libraries for common patterns (message preprocessing, context injection, response formatting)
• Implement conversation versioning, rollback capabilities, and testing frameworks (unit, integration, E2E)
• Design conversation state schemas with proper reducers for message history, user context, and session metadata

Real-Time Streaming & Backend Development:
• Build production FastAPI applications with async I/O for PostgreSQL, Redis, and LLM streaming responses
• Implement Server-Sent Events (SSE) streaming protocols for progressive LLM response rendering
• Design RESTful APIs for chat sessions, message history, conversation retrieval, and analytics
• Develop Postgres-based conversation persistence using PostgresSaver for stateful multi-turn interactions
• Handle concurrent async chat sessions with proper connection pooling, timeout handling, and resource management
• Containerize chat services using Docker and deploy via CI/CD pipelines with health checks and monitoring

LLM Integration & Prompt Management:
• Integrate cloud LLM providers (AWS Bedrock, Azure OpenAI, Anthropic Claude) with streaming support
• Design and maintain YAML-based prompt management systems with versioning, templating, and dynamic compilation
• Implement conversation memory patterns (sliding window, summary-based, selective retrieval) for context management
• Build response transformation pipelines converting LLM outputs into structured chat messages and UI-ready formats
• Develop hybrid chat patterns combining LLM reasoning with rule-based logic, validation, and business constraints
• Optimize token usage through prompt compression, message truncation strategies, and caching mechanisms

Agentic AI Support & Tool Integration:
• Support autonomous agent workflows by building chat interfaces for tool-calling and multi-step reasoning
• Implement Model Context Protocol (MCP) integrations enabling agents to call external data sources and APIs
• Design conversation flows handling agent reasoning traces, intermediate steps, and progressive disclosure of results
• Build UX patterns for human-in-the-loop agent interactions (user confirmations, feedback collection, corrections)
• Collaborate with agentic AI engineers to define chat-to-agent handoff protocols and state sharing mechanisms
• Develop streaming interfaces rendering agent tool calls, thoughts, and final answers in real-time

Observability, Security & Enterprise Integration:
• Integrate Langfuse or similar observability platforms for conversation tracing, LLM telemetry, and quality metrics
• Build monitoring dashboards tracking chat session metrics (latency, success rates, user satisfaction, token usage)
• Implement enterprise security patterns: LDAP/SSO authentication, row-level access control, PII filtering
• Develop audit logging for conversation history, user actions, and system decisions for compliance
• Create evaluation frameworks measuring conversation quality, coherence, factual accuracy, and user intent matching

Cross-Functional Collaboration & Mentorship:
• Partner with UX designers to translate conversational flows into intuitive chat interfaces
• Collaborate with frontend engineers on SSE consumption, message rendering, and real-time UI updates
• Work with data engineers to integrate chat workflows with data sources, caching layers, and analytics pipelines
• Mentor junior engineers on async Python patterns, LangGraph state management, and LLMOps best practices
• Participate in architecture reviews and contribute to documentation, runbooks, and knowledge sharing

 

Desired Experience* | Required:
Required:
• Bachelor's or Master's degree in Computer Science, AI, Data Science, Engineering, or related field
• 7+ years of software engineering experience with 2+ years building production LLM-powered conversational applications
• Expert-level async Python with LangGraph (state machines, message reducers, checkpointing), FastAPI, PostgreSQL, and Redis
• Production integration with cloud LLM providers (AWS Bedrock, Azure OpenAI, Anthropic, OpenAI) with streaming support
• Built multi-turn chat applications with Server-Sent Events (SSE), conversation state persistence, and prompt engineering at scale
• Docker containerization, Git workflows, and cloud platform experience (AWS/Azure/GCP)

Critical Conversational AI Experience:
• Designed conversation state schemas with message history management and context windowing
• Implemented streaming real-time LLM responses with proper error handling and progressive disclosure
• Understanding of conversational UX patterns and familiarity with tool-calling protocols (MCP, function calling)

Ideal Candidates Would Also Have:
• Potstgres checkpointing for LangGraph, observability platforms (Langfuse, LangSmith), or AWS Bedrock/Secrets Manager experience
• Prior experience in regulated industries (pharma, healthcare, finance) or global life sciences/GPS functional areas
• Experience working with globally distributed teams and strong communication skills across technical/non-technical audiences

Ideal Candidates Would Also Have:

• Prior experience in global life sciences especially in the GPS functional area will be a plus

• Experience working internationally with a globally dispersed team including diverse stakeholders and management of offshore technical development team(s)

• Strong communication and presentation skills

• Strong communication and presentation skills

If you come across a role that intrigues you but doesn’t perfectly line up with your resume, we encourage you to apply anyway. You could be one step away from work that will transform your life and career.

Uniquely Interesting Work, Life-changing Careers
With a single vision as inspiring as “Transforming patients’ lives through science™ ”, every BMS employee plays an integral role in work that goes far beyond ordinary. Each of us is empowered to apply our individual talents and unique perspectives in a supportive culture, promoting global participation in clinical trials, while our shared values of passion, innovation, urgency, accountability, inclusion and integrity bring out the highest potential of each of our colleagues.

On-site Protocol

BMS has an occupancy structure that determines where an employee is required to conduct their work. This structure includes site-essential, site-by-design, field-based and remote-by-design jobs. The occupancy type that you are assigned is determined by the nature and responsibilities of your role:

Site-essential roles require 100% of shifts onsite at your assigned facility. Site-by-design roles may be eligible for a hybrid work model with at least 50% onsite at your assigned facility. For these roles, onsite presence is considered an essential job function and is critical to collaboration, innovation, productivity, and a positive Company culture. For field-based and remote-by-design roles the ability to physically travel to visit customers, patients or business partners and to attend meetings on behalf of BMS as directed is an essential job function.

Supporting People with Disabilities

BMS is dedicated to ensuring that people with disabilities can excel through a transparent recruitment process, reasonable workplace accommodations/adjustments and ongoing support in their roles. Applicants can request a reasonable workplace accommodation/adjustment prior to accepting a job offer. If you require reasonable accommodations/adjustments in completing this application, or in any part of the recruitment process, direct your inquiries to adastaffingsupport@bms.com. Visit careers.bms.com/eeo-accessibility to access our complete Equal Employment Opportunity statement.

Candidate Rights

BMS will consider for employment qualified applicants with arrest and conviction records, pursuant to applicable laws in your area.

If you live in or expect to work from Los Angeles County if hired for this position, please visit this page for important additional information: https://careers.bms.com/california-residents/

Data Protection

We will never request payments, financial information, or social security numbers during our application or recruitment process. Learn more about protecting yourself at https://careers.bms.com/fraud-protection.

Any data processed in connection with role applications will be treated in accordance with applicable data privacy policies and regulations.

If you believe that the job posting is missing information required by local law or incorrect in any way, please contact BMS at TAEnablement@bms.com. Please provide the Job Title and Requisition number so we can review. Communications related to your application should not be sent to this email and you will not receive a response. Inquiries related to the status of your application should be directed to Chat with Ripley.

R1599643 : Manager AI Engineering