Senior AI Platform Engineer (Cloud-Native & Agents)

Bristol Myers Squibb EU Policy

Seattle (WA)

On-site

USD 151,000 - 183,000

Full time

4 days ago

Be an early applicant

Application generator

A complete application in a minute — tailored resume and cover letter, ready to send.

Get past ATS filters

Benefits offered by this job

Flexible Time Off

Medical Insurance

Dental Insurance

Vision Insurance

Retirement Plan

Company Holidays

Paid Vacation

Job summary

Bristol Myers Squibb is seeking a Senior AI Application Engineer for its Seattle-area AI Venture Studio delivery team. You will be hands-on, building secure cloud-hosted applications, designing APIs and agent runtimes, and delivering MVPs across six two-week sprints in an agile model.

You will work with LangGraph, FastMCP, OpenSearch, AWS services, and frontier LLMs to enable enterprise-scale AI pipelines, context management, and secure, observable software that aligns with pharma-wide needs.

Qualifications

Bachelor's or higher degree in Computer Science, Engineering, Science, or a related field.
5+ years of experience in software engineering, cloud engineering, platform engineering, or backend application development with increasing responsibility.
Hands‑on experience building cloud-native applications on AWS; familiarity with services such as S3, RDS/PostgreSQL, Athena, ElastiCache/Redis, OpenSearch, Fargate, Lambda, IAM, and VPC patterns.
Strong proficiency in Python, FastAPI, TypeScript/Node, or comparable backend application frameworks.
Experience with containers, CI/CD, GitHub-based workflows, automated testing, environment configuration, and infrastructure-as-code such as Terraform, AWS CDK, or CloudFormation.
Experience building LLM, RAG, or agentic AI applications using frameworks such as LangGraph, LangChain, PydanticAI, Claude Agent SDK, or similar tools.
Familiarity with MCP/FastMCP, read-write-search APIs, permissioned markdown/YAML stores, vector databases, knowledge graphs, session/state management, structured output validation gates, and evaluation-driven development.
Experience with SQL, semantic layers, data warehouse context, query history, and systems that translate LLM-derived meaning from unstructured scientific or operational sources into governed data/context layers.
Experience building sandboxed execution, data branching, provenance, version control, audit, and access-control patterns for agentic or data-intensive applications.
Practical experience integrating with model providers and a variety of approved frontier LLM models through enterprise AI services such as OpenAI, Anthropic, Gemini, AWS Bedrock, or similar approved channels.
Effective use of coding agents or AI-assisted development tools such as Claude Code, Codex, Gemini CLI, GitHub Copilot, or similar tools.
Excitement for experimenting with the latest AI tools and technologies while turning frontier prototypes into reliable foundations that help discover, develop, and deliver innovative medicines.
Curious and inquisitive mindset, with strong communication skills and comfort operating in fast-moving, cross-functional agile teams.

Responsibilities

Design, build, and operate backend services, APIs, and application components that power AI Accelerator products.
Develop Python/FastAPI, TypeScript/Node, or similar services that integrate LLM APIs, retrieval systems, workflow engines, and internal enterprise systems.
Execute AI Accelerator cycles of six two-week sprints over a 12-week cycle by developing, testing, and validating cloud and agentic AI product increments.
Develop MCP-accessible services that allow approved agents to read, write, search, and maintain structured (e.g. markdown/YAML) knowledge assets.
Build MCP/FastMCP read-write-search APIs, permissioned knowledge stores, version control, audit trails, access controls, and integrations with AWS-native storage and identity patterns.
Implement secure application patterns for authn/authz, BMS SSO, BMS Cloud Creds, secrets management, auditability, input validation, and safe service boundaries.
Partner with frontend engineers to define clean API contracts, streaming response patterns, error handling, and service-level behaviors for AI-powered user experiences.
Build and host agentic workflows using LangGraph, including workflow state, multi-agent orchestration, tool execution, fan-out/fan-in patterns, and durable checkpoints.
Develop MCP tool integrations and FastMCP servers that allow agents to use governed enterprise capabilities safely and consistently.
Implement retrieval, memory, and context services using AWS-aligned data stores such as S3, Athena, PostgreSQL/RDS, ElastiCache/Redis, OpenSearch, Amazon S3 Vectors, and Amazon Neptune.
Build and evolve the semantic layer for SQL and other natural-language-to-code generating agents, enabling analytical questions to be grounded in query history and warehouse context.
Package reusable deployment patterns, starter kits, and golden paths for AWS Fargate, serverless services, containers, and production-adjacent AI applications.
Create and maintain CI/CD pipelines, environment configuration, automated tests, infrastructure-as-code, and release processes for cloud AI applications.
Instrument reliability, latency, cost, usage, tracing, and model/agent behavior using LangSmith or similar tools.
Embed automated quality gates, security scans, regression tests, structured output validation gates, and guardrail checks into delivery pipelines.
Build sandboxed agent execution environments with recoverable transformations, provenance, and audit workflows.
Demonstrate MVP progress through bi-weekly demos and technical updates, tracking platform performance, reliability, cost, security, and business-value signals.
Continuously improve shared platform patterns based on lessons learned across pods and evolving enterprise standards.