Overview
Build agentic AI solutions that Finance and Global Business Services (GBS) teams use every day, taking them from discovery through prototype, MVP, production handoff and ongoing support, within the F&GBS Technology & AI Solutions Capability Center in Chennai.
What you'll do
- Work in person in Chennai while embedding virtually with an assigned Finance or GBS team, learning the work through regular calls, screen-share walkthroughs and process reviews.
- Run discovery with business users and process owners across regions, including the US and Europe.
- Translate business pain points into scoped use cases with success measures agreed with the business owner.
- Work from Product Owner priorities and specifications, and shape them into buildable scope with the business owner.
- Push back when a request does not match the underlying problem or is exposed to technical issues and limitations.
- Build agentic AI solutions end to end, including agents and tools on Azure AI Foundry, Vertex AI, other market-leading runtimes and Microsoft Agent Framework, LangGraph, LangChain, MCP tool integrations, retrieval with Azure AI Search, document extraction with Document Intelligence, and the interfaces users touch.
- Deliver working prototypes in days and MVPs in weeks, and iterate directly with end users.
- Write production-quality code in Python and TypeScript/JavaScript, with tests, version control and code review.
- Use AI coding agents in your daily workflow and review AI-generated code to the same standard as your own.
- Build evaluation sets that prove accuracy on real finance data before release, and run them as regression tests on every change.
- Connect solutions to enterprise Finance and GBS data and systems through governed MCP tools (or API) and certified data products, such as SAP S/4HANA, Snowflake (including Cortex), Databricks, SAP Ariba and Business Network, Oracle EPM, Microsoft 365 and other enterprise platforms.
- Write and tune SQL, and use and extend the shared F&GBS ontology, semantic views and business rules instead of hard-coding logic in prompts.
- Follow data access, entitlement and classification rules for financial and personal data.
- Design solutions that hold up to SOX, Internal Audit, data privacy (GDPR, India DPDP Act) and EU AI Act review.
- Record agent evidence, approvals and actions in the decision ledger, and build human review steps where AI outputs drive financial decisions.
- Apply Pfizer responsible AI safeguards: data access controls, prompt injection defenses and output monitoring.
- Document design decisions, data lineage and model behavior for control owners.
- Own production readiness and ongoing support, including LLMOps practice, deployment pipelines, monitoring, incident response and run costs.
- Establish support ownership, runbooks, service expectations, escalation paths, knowledge transfer and root cause analysis for production solutions.
- Deploy through approved pipelines on Pfizer cloud and GenAI platforms, and set up logging, tracing, monitoring and alerting with Azure Monitor, Application Insights and Foundry Observability.
- Contribute reusable components and patterns to the team's shared library, including the shared F&GBS Data & AI Foundation (ontology, rules engine, MCP tool library and decision ledger).
- Track usage, accuracy and hours saved, and report results to the business owner and FDE leadership.
- Train users and build champions in the business so adoption sticks.
What you'll need
- Bachelor's degree in Computer Science, Engineering, Data Science or a related field, with [6+] years of software or data engineering experience.
- 3+ years building LLM-based applications, including RAG, tool calling or agents, with at least one solution in use by business users.
- Experience evaluating and testing LLM output quality.
- Hands-on Python coding at a production standard.
- Working knowledge of TypeScript/JavaScript and a modern web framework such as React.
- Strong SQL and experience working with enterprise data platforms.
- Hands-on experience with Azure (preferred) or AWS cloud services, API design and CI/CD.
- Track record of working directly with business users to define and deliver solutions.
- Clear written and verbal communication with non-technical stakeholders.
Nice to have
- Master's degree in a technical field, or a finance/accounting background combined with engineering experience.
- Working knowledge of finance processes such as record-to-report, order-to-cash, procure-to-pay or FP&A.
- Experience with SAP data structures, Snowflake (including Cortex), or the Azure AI stack (AI Foundry, AI Search, Document Intelligence, Azure ML).
- Experience with agent orchestration frameworks (e.g., Microsoft Agent Framework, LangGraph), MCP and vector search.
- Exposure to ontologies, knowledge graphs or rules engines.
- Experience in a regulated industry, with exposure to SOX or IT general controls.
- Relevant certifications (AWS or Azure Developer or Solutions Architect, or other market-leading runtimes).
Details
- Location: Chennai, India, in person, with virtual embedding in Finance or GBS teams.
- Regular overlap with US and European time zones is required, including some early-evening IST calls.
- Must be able to travel to Pfizer offices and partner sites when required, including occasional international travel.
- Work can be demanding at times, including support during period-end close or response to production incidents.
Read the full description and apply on the company’s own careers page.