Junior AI Engineer

We help the world run better At SAP, we keep it simple: you bring your best to us, and we'll bring out the best in you. We're builders touching over 20 industries and 80% of global commerce, and we need your unique talents to help shape what's next. The work is challenging – but it matters. You'll find a place where you can be yourself, prioritize your wellbeing, and truly belong. What's in it for you? Constant learning, skill growth, great benefits, and a team that wants you to grow and succeed.

We are developing an agentic AI platform for Cloud Operations: autonomous agents that classify alerts, run operational playbooks, investigate incidents, and progressively refine their own skill library. The platform is organised into two layers — a control plane covering agent registration, skill governance, and human-in-the-loop orchestration, and the agents themselves, which execute on SAP BTP / Kyma.

This position lives at the boundary between AI engineering and platform engineering. We are hiring someone who ships agents that operate reliably in production — not proofs of concept.

What You'll Build

Designing and delivering LLM-driven agents based on skill-routing patterns (LangGraph, Claude API / Anthropic SDK, or comparable stacks)

Curating and evolving skill library — self-contained skill definitions with clearly declared tool bindings, trust levels, and confirmation checkpoints

Wiring agents into the systems already in production use: ServiceNow, alerting pipelines, Kyma / BTP APIs, and our internal knowledge sources

Contributing to shared agent memory and the feedback loops that lift skill quality over time

Packaging, deploying, and operating agents on SAP BTP / Kyma

Performing Agent testing and ensure Agents are production-ready, including authorizing tests that assess decision quality, not merely whether a task ran to completion

Results and Outcomes

Delivers agents and skills within an established architecture autonomously

Performs comprehensive testing on agents they build

Effectively works in accordance with established compliance, security and governance requirements

Understands ITSM and ITIL practices in the context of Cloud Operations and has them in mind to ensure Agent RoI, performance and relevance

Mindset

You focus on efficiency, value add and customer-centric experience

You actively communicate and collaborate with AI CoE and relevant teams to ensure compliance and effectiveness of AI Agents developed

You document your work — agent designs, skill contracts, and the reasoning behind decisions

You reason in terms of boundaries and contracts: what an agent is authorized to do unattended, what it must propose for approval, and what it must never do

You embed input from Operation team testing in Agent workflows prior implementation

You continuously work to develop a skill definition and initiate data-driven improvements over time

What You Bring

Essential

Strong command of Python (primary language for agent work), with working fluency in TypeScript / Node

Practical experience building LLM applications that go well beyond prompt crafting — including tool invocation, structured output handling, and agent control loops

A clear grasp of agentic design patterns: skill routing, human-in-the-loop workflows, and state-machine-driven execution

Working familiarity with at least one agent framework — LangGraph, LangChain, AutoGen, CrewAI, or an equivalent

Comfort with containers and foundational Kubernetes concepts, since agents run on SAP BTP / Kyma

A sharp production-safety focus: recognizing the failure modes that matter — hallucination, tool misuse, runaway loops, breaches of trust boundaries

Strongly Preferred

Hands-on exposure to SAP BTP, the CAP framework, or enterprise integration patterns.

Familiarity with ITSM and ITIL concepts — ServiceNow, the incident lifecycle, SLA management — since this is the domain in which the agents