AI/ML Engineer - Azure GenAI & Vector Search

AI/ML Engineer - Azure GenAI & Vector Search

Warsaw OR Krakow - 1 day per week on site

Contract: 6-12 months (B2B/Freelance)
Location: Remote, with occasional on-site visits in Poland (Krakow or Warsaw)
Language: English
Environment: Agile/Hybrid Agile

Role Overview
We are looking for an AI/ML Engineer to design and operate production-grade AI services built on Azure. The focus is on building semantic indexing, vector-based retrieval, data ingestion pipelines, and multimodal preprocessing to power knowledge and search systems. You'll collaborate closely with Back End, data, and Front End teams to bring models into production securely and reliably.

Key Responsibilities

• Build and maintain Azure-based AI components, including vector stores, semantic indexes, and retrieval services
• Develop embeddings, search strategies, and high-performance ANN workflows
• Implement and optimize data pipelines for ingestion, preprocessing, and transformation
• Integrate AI features into existing applications and ensure smooth end-to-end delivery
• Continuously evaluate, tune, and improve model accuracy and performance
• Provide technical leadership, contribute to architectural decisions, and support team standards
• Mentor less experienced engineers and share ML/AI best practices
• Work with OCR and image-processing flows to support multimodal search and extraction
• Contribute to AI agent design and data architecture for automation use cases

Required Skills & Experience

• 3-6+ years hands-on experience building ML/AI solutions, with at least 2 years using modern LLMs or vector-search ecosystems
• Strong practical experience with Azure: Azure OpenAI, Cognitive Search or vector indexing, Data Factory, Databricks, AKS/ACA, Key Vault, App Configuration
• Proven experience creating embeddings, designing semantic indexes, and working with similarity search/ANN libraries
• Solid programming skills in Python and familiarity with ML frameworks (transformers, sentence-transformers, scikit-learn, PyTorch/TensorFlow)
• Advanced prompt engineering skills, including templates, safety/guardrails, and production prompt workflows
• Data engineering experience (ETL/ELT pipelines, batch/stream processing, data quality, metadata)
• Knowledge of OCR/image processing tools (Tesseract, Azure Form Recognizer, OpenCV or similar)

Nice to Have

• Experience with Semantic Kernel, agent frameworks, or multi-agent orchestrations
• Hands-on knowledge of RAG architectures and hybrid search (BM25 + semantic)

If this is of interest - please apply using your latest CV and let's discuss further!