Location: Hybrid option at our Stockholm HQ
Reports to: CTO
What You’ll Do
- Build and deploy LLM‑powered features: retrieval‑augmented generation, autonomous agents, prompt‑chains.
- Fine‑tune open‑weight models and evaluate commercially‑hosted LLM APIs (GPT‑4o, Claude, Gemini).
- Design robust knowledge‑graphs and vector‑search pipelines to surface supplier data.
- Collaborate with Product and Data Science to translate fuzzy ideas into shippable AI products.
- Promote agentic coding and AI‑first development practices across engineering teams.
Tech You’ll Touch
LangChain · Python · Kubeflow · Vertex AI / Bedrock / Anthropic APIs · Vector DBs (PGVector, Weaviate) · BigQuery · Looker · Kubernetes · Terraform · Elastic
What We’re Looking For
- 3+ years building ML/AI products, with recent focus on large‑language‑model applications.
- Strong Python skills and familiarity with modern ML stack (PyTorch, HuggingFace, TensorFlow-lite).
- Practical knowledge of vector search, embeddings, and retrieval systems.
- Experience shipping models to production and monitoring real‑world performance.
Nice‑to‑Have
- Contributions to open‑source LLM tooling or research publications.
- Experience with autonomous agent frameworks or reinforcement‑learning‑with‑feedback.
Why Join Us
- Direct access to OpenAI, Claude, and Gemini models for experimentation.
- Opportunity to shape an AI‑first culture and influence product roadmap.
- Hybrid/remote flexibility and competitive salary + equity package.
Application Process
Screening call → Technical deep‑dive → Take‑home prompt‑engineering task → Team fit → Offer.
