Agentic AI Developer (Forward Deployed)
Choosing Capgemini means choosing a company where you will be empowered to shape your career in the way you’d like, where you’ll be supported and inspired by a collaborative community of colleagues around the world, and where you’ll be able to reimagine what’s possible. Join us and help the world’s leading organizations unlock the value of technology and build a more sustainable, more inclusive world.
Your role
As an Agentic AI Developer you will co-create and deliver intelligent AI solutions together with a collaborative and diverse team which are embedded with our clients, designing, developing, and deploying end-to-end agentic AI systems that are intelligent, autonomous solutions.
This role combines AI engineering, software development, and system integration expertise to build scalable, production-ready solutions leveraging LLMs, vector databases, orchestration frameworks, model context protocols (MCP), tool use, evaluation frameworks, and modern web technologies.
You will work close to the client, shaping use cases with the people who will use them, and iterating on working software rather than waiting for a specification. You will collaborate in cross-functional teams with data scientists, developers, and product specialists, where knowledge-sharing and support are part of everyday work.
In this role you will play a key role in:
- GenAI & Agent Engineering: architect and implement agentic workflows and multi-agent orchestration using GenAI models (e.g., Claude, GPT, Gemini, Kimi): agent roles, memory systems, tool integrations, MCP-based integrations, and inter-agent communication protocols.
- Work with orchestration platforms such as LangGraph, Agent SDK, or Google ADK and cloud AI platforms (AWS Bedrock, Vertex AI, Azure AI Foundry, Agent Builder), optimizing agent workflows for reasoning, planning, tool use, reliability, and cost efficiency.
- Prompt & Context Engineering: design and iterate high-quality prompts for agentic tasks, with prompt templates, context management strategies, and RAG pipelines using vector databases (e.g., Qdrant, Pinecone, Weaviate). Improve accuracy, latency, safety, and cost through prompt optimization, retrieval quality assessment, grounding validation, and automated evaluation frameworks.
- Backend Development: build and maintain backend services and RESTful APIs supporting agent orchestration and user interfaces. Develop microservices using Python (FastAPI, Flask) or Node.js with the scalability and security to integrate with cloud-native services such as BigQuery, Pub/Sub, Firestore, messaging queues, and enterprise data platforms.
- Cloud & DevOps: deploy and monitor GenAI applications on AWS, Azure, or GCP using CI/CD pipelines, infrastructure-as-code (Terraform), and containerization (Docker, Kubernetes). Implement logging, tracing, observability, evaluation pipelines, and performance monitoring for agent-based systems.
- Client Delivery: run discovery and feasibility assessments that turn client ambitions into scoped use cases. Build prototypes fast, put them in front of real users, and navigate the realities of enterprise delivery like data access, legacy integration, security reviews. Document what you build and enable client teams to run it.
Your profile
- Experience with Python, SQL, JavaScript, TypeScript, or similar programming languages.
- Experience with GenAI frameworks such as LangGraph, Anthropic Agent SDK, OpenAI Agents SDK, or similar.
- Strong understanding of LLM orchestration, tool calling, RAG, MCP, prompt engineering, evaluation methodologies, and agentic workflows.
- Experience with backend development (FastAPI, Flask, Node.js), cloud deployment, and cloud AI platforms (AWS Bedrock, Vertex AI, Azure AI Foundry).
- Experience with AI observability, evaluation frameworks, prompt/version management, and production AI operations.
- Comfortable engaging senior stakeholders and working in ambiguous environments where requirements are not yet fully defined.
- Degree in Engineering, Computer Science, IT, or a related quantitative field.
- Fluent in English.
If you also have experience with AI observability, evaluation, governance, safety, compliance and speaks Swedish fluent it's a strong plus.
Stockholm, SE Göteborg, SE