We're Nagarro.
We are a Digital Product Engineering company that is scaling in a big way! We build products, services, and experiences that inspire, excite, and delight. We work at a scale across all devices and digital mediums, and our people exist everywhere in the world (18000+ experts across 36 countries, to be exact). Our work culture is dynamic and non-hierarchical. We are looking for great new colleagues. That is where you come in!
Requirements
- 7+ years designing and operating distributed systems in production, including ownership through real incidents.
- Production coding experience in Python and at least one other language such as TypeScript, Java, .Net or Go.
- 2–3 years shipping LLM-based or agentic systems to real users, with clear understanding of failures and iterations.
- Experience building evaluation frameworks for AI systems and using them to gate releases.
- Demonstrated judgment in choosing deterministic solutions over agents when appropriate.
- Experience introducing AI-assisted development practices to engineering teams with sustained adoption.
- Hands-on experience with agent orchestration frameworks, MCP for tool integration, and retrieval pipelines (RAG).
- Ability to design retrieval layers including document ingestion, chunking, embeddings, vector and hybrid search, reranking, and knowledge graphs.
- Experience optimizing assistant performance and cost for concurrent users using caching, streaming, model routing, and access-aware retrieval.
- Deployment and operation experience on at least one major cloud platform with containers, Kubernetes, and CI/CD pipelines.
- Experience navigating enterprise controls including security review, access management, and audit requirements.
- Experience running agents with governed tool access, logged actions, and budget-constrained costs.
Responsibilities
- Design and build agentic workflows in client environments including multi-agent pipelines, MCP servers, agent skills, and evaluation suites.
- Rework client software development lifecycles so agents participate in planning, coding, testing, and review with clear human oversight.
- Define verification approaches for agent output including evaluations, guardrails, observability, and deterministic coverage checks.
- Take agentic systems from prototype to production on client cloud infrastructure with containerized deployment, CI/CD, monitoring, cost controls, and rollback.
- Pair with client engineers and coach them until AI-assisted development becomes their standard practice.
- Make and own architectural decisions and communicate them clearly to both engineers and executives.
- Convert successful patterns into reusable playbooks and accelerators for broader organizational reuse.
- Mentor architects and engineers in agentic AI practice.
- Lead the technical side of RFP responses, proofs of concept, and effort estimates for agentic AI work.