Jobs in Germany
We’re looking for an AI Engineer to build and iterate on the core agentic systems behind our claims automation engine. You’ll work on AI agents, retrieval pipelines, evaluation frameworks, and custom tooling to ensure reliability, auditability, and performance of our AI systems in production.
This is not just another ML role—you’ll be working at the intersection of research and real-world impact, shipping systems that solve messy, high-stakes tasks in production, inside a highly regulated domain.
If you love building agentic AI systems, hacking on retrieval pipelines, running evals at scale, and making AI safe and usable in the wild—this role is for you.
Who We Are
We are Inca – a venture-backed startup on a mission to automate insurance claims through Agentic AI.
- 💥 We’ve just closed one of the largest and most competitive InsurTech Pre-Seed rounds in Europe – backed by top-tier VCs.
- 🚀 We bring Agentic AI to insurance industry with a no-bullshit approach — delivering cutting-edge results, where others overpromise.
- 🔥 The market is on fire: we’re flooded with inbound interest from insurers eager to transform their claims processes.
- 🧠 Our team combines former BCG Project Leaders, industry- and AI-experts. Together, we’re redefining how insurances are operating in Europe.
What we are Building
At INCA, we’re building the engine behind the future of insurance. Our hyper-modular Agentic AI platform is redefining how automation scales in heavily regulated enterprise environments. Powered by the latest frameworks in agentic computing, our system isn’t just smart — it’s redefining the speed, service quality, and fairness of insurance.
- 🧠 At the core: a modular Agentic Engine that commands specialized agent teams, each built for mission-critical tasks across the insurance claims lifecycle.
- 🛠️ Each agent is equipped with best-in-class tools, integrated seamlessly — from data retrieval to decision support to payout optimization.
- 🎯 The result? A powerful, compliant, and highly adaptive AI system that sets a new bar for enterprise-ready Agentic AI.
Tasks
You’ll work side-by-side with our CPO — but that’s not all. You’ll also have direct mentorship from our Head of Engineering. With 15+ years of startup experience — including at BCG Digital Ventures, SoundCloud, and Issuu — he’s scaled products from early MVP to millions of users, built world-class teams, and seen what it takes to win. You’ll learn how to build right — and build fast.
- Build Agentic Systems: Use frameworks like LangChain, LangGraph, or AutoGen to build robust AI agents that perform structured tasks with autonomy and reasoning
- RAG Pipelines: Design and deploy retrieval-augmented generation systems to ensure accurate, grounded outputs from LLMs
- Evaluate & Iterate Agents in Practice: Build evaluation loops using automatic metrics (BLEU, ROUGE, BERTScore), self-consistency, chain of doubt, LLM-as-judge, or custom classifiers
- Production Integration: Work closely with backend and infra teams to integrate AI models into real products
- Own AI Quality: Create tools and dashboards, CI hooks, or human-in-the-loop flows that ensure reliable auditable & explainable AI behavior at scale
Requirements
- 2–5 years' experience in AI/ML engineering**,** with a focus on LLMs, RAG, or NLP systems in a high-paced environment (e.g., start-up, scale-up)
- Solid Python skills and hands-on with modern AI tooling (e.g., RAG-optimization, vector data bases, evaluation tools etc.)
- Familiarity with prompt engineering, embeddings, and evaluating LLM outputs in production
- Proven track record shipping in fast iteration loops with unsolved challenges in highly dynamic environment
- Proactive, curious, and user-oriented—you want to build highly autonomous, scalable and auditable agentic systems with the best tools and frameworks in the market
Bonus Points For
- Experience with AWS, GCP, or Azure for model deployment or vector storage - we will set-up AI agents in a multi-cloud environment
- Familiarity with LLM finetuning or low-rank adaptation (LoRA)
- Experience with insurance, regtech, or other high-compliance domains
Your First 30 Days
- 🤖 Ship an Agent: Build or refine an internal AI agent for a specific task (e.g. claim triage, document QA)
- 🔍 RAG Evaluation: Benchmark and compare two RAG pipelines using automatic and LLM-based evals
- 📈 Build Eval Dashboard: Make model quality and failure cases transparent across the team
- 🛠 Optimize Retrieval: Improve embedding relevance or chunking strategies for better grounding
Benefits
Why Join Us?
- 🤖 AI That Ships: Build real-world LLM systems—not just research prototypes
- 🔍 Agentic Infrastructure: Own the logic, structure, and behavior of AI agents in production
- 🛠 Cutting Edge Stack: Use the best tools in Agentic AI and help us push them further
- 🧠 High Autonomy: Drive both research and implementation decisions alongside experienced teammates
- 🪙 Equity + Competitive Salary: Strong base + ESOPs
- 🧳 Location: Offices in Berlin or Munich (3+days per week), frequent team events
- 🧳 Holiday: 30 days of vacation & up to two weeks “work-from-everywhere”
- 📈 Culture: We are builders and are dedicated to building an open, inclusive and driven culture where engineers thrive - professionally and personally.
