Location: Berlin, Germany (office first)

Start date: Immediately

Contact: Leo Wossnig (l[email protected])

Intro

Forgent AI is on a mission to build AI products for better public procurement. We are looking for a talented Product Engineer to join our founding technical team in Berlin. You'll play a crucial role in shaping our product and technology and building reliable, impactful software for a domain where it truly matters.

The role

As an AI Engineer at Forgent AI, you will design, build, and optimize the sophisticated AI systems that differentiate our platform. This is a unique opportunity within an early-stage, mission-driven company to significantly shape our AI strategy and technical direction. You will tackle complex challenges applying your expertise in LLMs and RAG systems, from prompt engineering to LLM fine-tuning. Working closely with founders and other engineers, you'll have substantial ownership over key AI components, driving innovation from experimentation through to production-ready systems, and helping establish our AI development culture and best practices.

Your day-to-day

• Design, implement, and evaluate state-of-the-art AI models and systems, with a specific focus on LLMs and vector search/retrieval techniques (e.g., using pgvector or dedicated services, RAG).
• Develop and refine methods for extracting structured information and relationships from unstructured text data.
• Collaborate closely with product engineers, platform engineers, and founders to integrate AI components into the architecture and ensure alignment with user needs.
• Take ownership of specific AI components, managing data preparation, driving experimentation cycles, prototyping new approaches, and rigorously evaluating model performance.
• Stay abreast of the latest advancements in relevant AI fields and proactively identify opportunities to apply novel techniques to enhance our product.
• Contribute to technical design discussions, share research findings and experimental results with the team, and help establish best practices for AI development, deployment, and monitoring.

You should apply if you

• Are a skilled AI Engineer with 3+ years of professional experience building and deploying sophisticated AI/ML systems into production environments.
• Deep hands-on expertise with LLMs including model orchestration, agentic frameworks (e.g. LangChain, LangGraph, or similar frameworks), MLOps infrastructure (e.g. evals data), and advanced prompt engineering techniques (e.g. zero shot prompting, CoT, etc.). This should include:
  • Deep understanding of models, their context windows, and model behavior, and limitations (including practical limitations such as API call limits),
  • Experience with building agentic applications,
  • Collecting eval data in production and leveraging evals for system optimization.
• Have strong practical experience implementing and optimizing RAG systems, including:
  • Deep understanding of state-of-the-art embedding (e.g. contextual embeddings), retrieval (e.g. hybrid search), and chunking techniques,
  • Ability to implement and optimize vector searches (e.g., using Qdrant, Pinecone, or Weaviate),
  • Deep knowledge of model orchestration approaches and related infrastructure challenges to build scalable LLM-based and agentic solutions (e.g. ability to leverage frameworks such as Temporal).
• Can demonstrate successfully translating complex requirements or research concepts into practical, working AI solutions that deliver tangible results.
• Are an excellent communicator (written and verbal English) who thrives on deep collaboration, including code reviews and asynchronous design discussions.