Capgemini · Posted 2w ago
AI Development FDE
Job relayed by Stapply Data, its source: you apply there.
Consultancy. Consultancy or IT services firm: the work usually happens on assignment at a client.
- Role
- Data science
- Source
- Stapply Data
Stack
- LLM
- Vector DB
- Python
- TypeScript
- Azure
- CI/CD
The posting
My Role Develop and integrate enterprise AI solutions, including GenAI apps, RAG pipelines, AI agents, copilots, automation workflows and reusable AI platform components. Build GenAI apps, RAG components, prompt flows, data connectors, APIs and AI agents. Design and develop responsive user interfaces using React, Next.js, TypeScript, HTML5, CSS3, and modern UI component libraries (Material UI, Tailwind CSS or similar). Implement vector indexes, embeddings, document processing and knowledge grounding using Azure AI services. Create PoCs, MVPs and reusable accelerators for client demonstrations and delivery mobilization. Support testing, evaluation, groundedness checks, performance tuning, monitoring and cost optimization. Document designs, deployment steps and reusable implementation patterns. Collaborate with consultants, product owners, architects, UX designers and governance specialists. Experience and Skills Required 5-8 years in software, data, AI, cloud development or automation. Hands-on work building GenAI, Agentic Harness, RAG, chatbot, AI/ML or enterprise integration solutions. Good Python and frontend capability plus APIs, SDKs and source control. Experience in agile teams and ability to communicate technical trade-offs clearly. Python, FastAPI or similar backend and frontend frameworks. Strong experience in Python and modern frontend development using React or Next.js. Azure OpenAI, Azure AI Foundry, Azure AI Search and Azure ML fundamentals. RAG pipelines, embeddings, vector search, prompt engineering and evaluation. Agent frameworks, orchestration, tool calling and workflow automation. APIs, databases, message queues, files and enterprise connectors. Git, CI/CD, containers, configuration and automated testing. AI quality: test datasets, regression checks, hallucination checks, groundedness, latency and cost. Responsible AI implementation: content filters, guardrails, logging and human review.