Field AI Engineer

Overview

Field AI Engineer on the Global Sales Systems and Capabilities team within CBI&T (Commercial Business Insights & Technology) in Hyderabad. The role builds AI-powered capabilities for the field force on Veeva Vault CRM, from LLM-based assistants and agentic workflows to the data pipelines and integrations that ground them in trusted commercial data. It is an applied AI engineering role, not deep ML research or model training.

What you'll do

• Design, build and ship AI/LLM-powered features for Vault CRM field users, such as pre-call insights, call note summarization, conversational assistants and next-best-action recommendations, using prompt engineering, retrieval-augmented generation (RAG), tool/function calling and agentic workflows.
• Build and maintain the data foundation for AI: cloud-based (AWS) batch and event-driven pipelines (Glue, Lambda, API Gateway, EventBridge) and integrations with Vault CRM (e.g., Vault REST APIs, Direct Data API) and related commercial systems that supply clean, timely and governed context to AI applications.
• Implement retrieval components, including document ingestion, chunking, embeddings, vector/hybrid search and metadata filtering, and tune them for relevance and accuracy.
• Integrate foundation models through enterprise platforms (e.g., Amazon Bedrock, Anthropic, Azure OpenAI) and expose AI capabilities through secure APIs, Model Context Protocol (MCP) servers and/or native Vault CRM extensions.
• Make evaluation part of the build: create test datasets, define quality metrics (accuracy, groundedness, hallucination rate) and run automated and human-in-the-loop evaluations before and after each release.
• Monitor AI applications in production (tracing, latency, token cost and user feedback) and continuously improve prompts, retrieval and model choices based on what the data shows.
• Apply responsible AI guardrails appropriate for a regulated life sciences environment, including PII handling, content safety, access controls and commercial compliance considerations, in partnership with security, privacy and AI governance teams.
• Bring a product mindset: partner with field users, business stakeholders and product managers to understand workflows, shape use cases, build quick prototypes to validate value, and define success measures such as adoption, time saved and user satisfaction.
• Stay current with fast-evolving AI tools and practices, use AI-assisted development tools effectively, and share learnings with the broader Field Integrations team.
• Participate in effort estimation, code reviews, CI/CD and production support, and maintain technical documentation such as design documents, deployment checklists and testing/validation records.

What you'll need

• Bachelor's degree in Computer Science, Information Technology, or related field.
• 2 to 4 years of IT experience in software development or data engineering.
• At least 1 year of hands-on experience building LLM/GenAI applications (production experience preferred; strong project or POC work will be considered).
• Strong programming skills in Python, with sound engineering practices: clean code, testing, version control (Git) and CI/CD.
• Good SQL skills and understanding of relational databases (preferably PostgreSQL); experience building and consuming RESTful APIs.
• Hands-on experience with AWS services such as Lambda, API Gateway, Glue, S3 and EventBridge.
• Practical experience with LLM APIs and at least one orchestration/agent framework (e.g., LangChain/LangGraph, LlamaIndex, Strands Agents, Bedrock Agents), including prompt engineering, structured outputs and tool/function calling.
• Understanding of RAG architecture, embeddings and vector stores (e.g., OpenSearch, pgvector).
• Familiarity with LLM evaluation and observability approaches and awareness of common failure modes such as hallucination, prompt injection and data leakage.
• Product mindset: ability to turn an ambiguous business problem into a well-scoped AI solution, explain trade-offs (quality, latency, cost) clearly to non-technical stakeholders, and iterate based on user feedback.
• Ability and willingness to learn new technologies quickly; proactive and self-motivated, with a strong sense of ownership and accountability.

Nice to have

• Experience with Veeva Vault CRM or Veeva CRM (Salesforce-based), Vault APIs, or other CRM platforms.
• Life sciences / pharma commercial domain knowledge (field force effectiveness, HCP engagement, compliance).
• Exposure to MCP, multi-agent patterns, or speech-to-text/voice AI use cases.
• Ability to build lightweight front-ends for AI prototypes (e.g., React).
• Experience with Docker and infrastructure-as-code (Terraform, CloudFormation or AWS CDK).
• AWS or GenAI certifications.

Details

• Location: Hyderabad, India.
• Work model: assigned based on the role's core responsibilities; see careers.bms.com/ways-of-working.