Generative AI Engineer (Product Intelligence)

Quick Facts

• Role: Generative AI Engineer (Product Intelligence)
• Focus: Productionizing generative AI for product data intelligence (LLMs, embeddings, scoring, evaluation)
• Location/Work style: Based in or near Kraków; ~60% onsite/local collaboration, rest remotely

Description

Join a generative AI team building a Product Intelligence Platform that automates product categorisation, enrichment, and matching at scale. You will run feasibility studies and proofs of concept, then take promising ideas into production by integrating and evaluating LLMs, embeddings, and scoring pipelines. Work mainly in Python on event-driven Kafka services, collaborating closely with product managers, data stewards, and engineers.

Responsibilities

• Lead generative AI feasibility studies and PoCs for new product intelligence capabilities; convert successful results into production designs
• Design and build GenAI solutions integrating large language models (Azure OpenAI GPT-4) via a GenAI Service facade, including prompt design, evaluation, and iteration
• Build and maintain embedding-based similarity and scoring pipelines using PostgreSQL with pgvector
• Contribute to LLM observability and evaluation practices (e.g., Langfuse), including regression testing and drift detection
• Support Human-in-the-Loop (HITL) tooling for data steward review and validation (100% HITL review before publication)
• Contribute to responsible GenAI governance (audit support and structured release criteria for prompt/model changes)
• Build event-driven Python services for high-volume data processing on Kafka
• Translate ambiguous business problems into clear technical designs with stakeholders
• Share GenAI knowledge, support team upskilling, and optionally represent the work externally

Requirements

• Solid, recent experience building production backend services in Python using a modern framework (FastAPI, Django, Flask, or similar)
• Hands-on experience with LLMs and prompt engineering in production or near-production
• Experience with embeddings and vector similarity (pgvector or equivalent vector database)
• Strong understanding of event-driven architectures and Apache Kafka
• Ability to tackle ambiguous, research-shaped problems via feasibility studies/PoCs and deliver production solutions
• Cloud experience, preferably Azure
• Software engineering best practices: CI/CD, testing, and code review
• Strong stakeholder communication skills
• Based in or near Kraków

Benefits

• Permanent contract from the start
• MacBook for work
• Learning opportunities (certified technical training and platforms like Udemy)
• Referral bonus
• Sports activities with a personal trainer in the office
• Additional 4 days of paid leave
• Up to 20% yearly salary bonus (individual and business performance)
• Private healthcare (LuxMed)
• Cafeteria & Multisport
• IP tax deductible costs
• Holiday support: expands holiday pool from 20 to 25 days for those not yet eligible for full entitlement