Data Scientist – RAG & Document Intelligence

Craftware is a technology company of over 500 experts, empowering large organizations to solve complex business challenges with modern IT solutions - from sales systems and automation to data platforms and AI. We operate where technology must be reliable, secure, and scalable. We deliver end-to-end projects: from analysis and architecture through implementation to development and maintenance. We are a trusted partner of industry leaders such as Salesforce, Veeva, UiPath, and Databricks.

Model: remote
Engagement: full-time

Role summary

You’ll join one of the most exciting AI initiatives at an international Consumer Health company: a platform that lets business users retrieve, synthesize, and act on knowledge locked inside complex enterprise documents, using plain language. This is a greenfield, research-meets-engineering role. You’ll design, experiment with, and implement a context-aware, multi-agent AI system, with real influence over model selection, retrieval strategies, and quality standards. Accuracy is the product here: you’ll own the quality bar and have the mandate to keep raising it for business users across commercial, marketing, product supply, and R&D.

Responsibilities

• Optimize RAG pipelines through experimentation: systematically test parsing and chunking strategies, embedding models, retrieval algorithms, re-ranking approaches, and context assembly techniques to maximize answer quality and minimize hallucination.

• Drive document intelligence: evaluate and benchmark document parsing solutions for PDFs, Word documents, presentations, scanned files, and mixed-format corpora, and extract structured meaning from noisy inputs, including tables, charts, and figures.

• Evaluate and compare models: design and run rigorous experiments comparing LLMs, embedding models, and retrieval models across accuracy, latency, reliability, and cost. Build evaluation datasets and define RAG-specific quality metrics.

• Apply NLP and prompt engineering: work on semantic similarity, named entity recognition, query rewriting, and multi-turn conversation handling. Craft, version, and systematically evaluate prompts and prompt chains.

• Balance quality and cost: track token consumption, latency, and retrieval quality across experiments, and make principled trade-offs between model capability and operational cost at enterprise scale.

Requirements

Must-have:

• Strong Python skills with hands-on experience in machine learning and generative AI workflows, including production-grade code

• Solid NLP background: text representation, semantic search, embeddings, and language model behaviour

• Deep hands-on experience designing and optimizing RAG pipelines over unstructured documents: chunking strategies, embedding models, hybrid search, re-ranking (e.g. LightRAG, LlamaIndex, LangChain)

• Experience with document parsing and noisy, multimodal document formats, including extracting information from tables, charts, and figures

• Experience with LLM evaluation frameworks and RAG-specific quality metrics (e.g. RAGAS, DeepEval, Langfuse)

• Familiarity with multi-agent AI frameworks such as LangGraph, LangChain, or Pydantic AI

• Experience with vector databases (e.g. pgvector, Pinecone, Qdrant, Weaviate)

• Experience with Azure cloud services (Azure Apps, Containers, Storage, AI Search, AI Foundry) and/or AWS equivalents

• Experience with Databricks (Delta Lake, Unity Catalog, MLflow)

• Strong analytical mindset with a structured, experiment-driven approach to problem solving

• Fluent English, both written and spoken

Nice-to-have:

• Hands-on experience with Databricks GenAI products: Vector Search, Agent Framework, Knowledge Assistance, Genie, Agent Bricks

• API development and integration skills (e.g. FastAPI)

• Knowledge of Model Context Protocol (MCP) for tool integration

• Understanding of knowledge management, taxonomy design, and metadata enrichment for enterprise document repositories

We offer:

• B2B contract (rate negotiated individually, up to 200 PLN net/h + VAT)

• Fully remote service delivery

• Broad range of projects (internal, international) - genuine variety of clients and tasks

• Budget for skills development and certifications as part of the collaboration

• Regular collaboration reviews and discussion of project scope

• Additional benefits available as part of the collaboration

• Networking and team-building events for project teams

Tech stack

• Python
• RAG
• NLP
• LLM
• DataLakes
• Azure Apps