GenAI Platform Engineer – LLM Gateway
Pfizer
·Today
- Location
- IND - Remote, India
- Workplace
- Remote
- Type
- Full-time
- Department
- Engineering
- Experience
- 9+ years
- Closing date
- Today
- Source
- Workday
Description
ROLE SUMMARY
The GenAI Platform Engineering team builds and runs the shared platform that lets Pfizer colleagues and applications use generative AI safely and at scale. At its core is the enterprise GenAI Gateway: a single, governed entry point to large language models from multiple providers, with self-service access, routing, usage and cost transparency, and observability built in. Around it, we deliver evaluation, observability and agentic tooling — MCP integrations, agent registries and reusable agent skills — for teams across the company.
We run the platform as a product: self-service over tickets, automation over manual approvals, small and frequent releases, and reliability we can measure. We are looking for a hands-on engineer to help evolve the Gateway and its ecosystem. You will keep it secure, fast and current in a fast-moving open-source landscape. You will work closely with a globally distributed team and with the teams that build on the platform.
This is an individual contributor role.
ROLE RESPONSIBILITIES
Platform engineering
- Design, build and operate features of the enterprise LLM gateway, mainly in Python:
- multi-provider model access
- routing and failover
- API key and access management
- budgets, usage metering and chargeback
- Integrate agentic capabilities into the platform with the right security controls. Examples include Model Context Protocol (MCP) servers, tool calling across providers, and agent-to-agent connectivity.
- Keep the platform’s open-source foundations current. Assess upstream releases, maintain and retire Pfizer-specific extensions, and deliver upgrades without regressions.
- Contribute to the platform’s admin and self-service experiences, including web UI, portals and developer documentation.
Quality, reliability and security
- Build and maintain automated quality gates in CI/CD. These include unit, integration, contract and performance/load tests that protect every release.
- Instrument services with metrics, traces and logs (e.g., OpenTelemetry, Langfuse). Use them to find and fix latency, error and cost issues.
- Take part in the support/on-call rotation. Lead root-cause analysis of production incidents, and turn the findings into lasting fixes and blameless postmortems.
- Apply secure-by-default engineering:
- identity and access controls (SSO, OAuth 2.0/OIDC, RBAC)
- secret handling
- dependency pinning and supply-chain hygiene
Enablement and collaboration
- Evaluate emerging GenAI capabilities hands-on, such as model routing, guardrails and agent registries. Turn the results into clear recommendations on whether to adopt them.
- Help consuming teams onboard and succeed. Write guides and runbooks, and create reusable assets such as code samples and AI agent skills.
- Work in the open. Make InnerSource contributions, review code, and engage with open-source communities and vendors by reporting issues and proposing fixes upstream.
BASIC QUALIFICATIONS
- Bachelor’s or Master’s degree in Computer Science, Computer Applications, Engineering or a related field, or equivalent practical experience.
- 9+ years of professional software engineering experience, including 3+ years building GenAI/LLM-based applications or platforms used in production.
- Strong Python skills for backend services: async programming, FastAPI or similar, REST APIs and streaming responses.
- Hands-on experience with LLM provider APIs (e.g., OpenAI, Azure OpenAI, AWS Bedrock), including tool/function calling and integrating multiple providers.
- Experience delivering and running services on AWS with containers and CI/CD pipelines (e.g., Docker, GitHub Actions), with automated testing built into the workflow.
- Proven ability to troubleshoot complex production issues and perform root-cause analysis.
- Clear written and verbal communication in English, and comfort working with distributed, cross-functional teams.
PREFERRED QUALIFICATIONS
- Experience with an AI/LLM gateway or API gateway, including routing, key management and cost tracking.
- Experience running and extending open-source software in production, including patches, upgrades, regression testing and upstream contributions.
- Knowledge of agentic AI patterns and tooling:
- MCP
- LangGraph/LangChain or similar frameworks
- multi-agent orchestration
- RAG and vector search
- Experience with performance/load testing and observability tooling (e.g., OpenTelemetry, Langfuse, Prometheus/Grafana).
- Experience with Kubernetes and Infrastructure as Code (Terraform).
- Frontend skills (React, TypeScript) for internal tools and portals.
- Daily use of AI-assisted development tools (e.g., GitHub Copilot, Claude Code) and interest in building reusable agent skills.
- Experience in pharmaceutical, healthcare or other regulated environments.
- Experience mentoring engineers or delivering technical enablement, such as workshops, documentation and onboarding.
- AWS certification.
WORK SCHEDULE
- Regular collaboration with colleagues in European time zones.
- Participation in the team’s support rotation.
Pfizer is an equal opportunity employer and complies with all applicable equal employment opportunity legislation in each jurisdiction in which it operates.
To learn more about acceptable and prohibited uses of AI during the recruitment process, please review our candidate AI-use guidelines available on Pfizer Careers.
Information & Business Tech