Principal AI Engineer

Svitla Systems Inc. is looking for a Principal AI Engineer for a full-time position (40 hours per week) in Ukraine, Europe. Our client is a middle-market private equity firm that gathers information and reports to enhance environmental and social impacts on the customer’s business.

We are seeking a visionary Principal AI Engineer to lead the development of the AI Platform. As a founding member of the AI Center of Excellence, you will lead by example to build a high-performance team. You will build sophisticated Agentic Systems and Adaptive RAG engines that conduct autonomous research to solve complex, real-world problems. You will be a key member of defining "how we do AI, setting the standards for engineering quality, security, and innovation.

Tech Stack:

• Languages: Python, TypeScript;
• AI Frameworks: LangChain/LangGraph, Pydantic AI, Databricks Mosaic AI;
• Models: Azure OpenAI, Open Source LLMs (Llama 3, Mistral, etc.)
• Infrastructure: Microsoft Foundry, Azure AI Search, Azure Container Apps, Delta Lake, Vector;
• Databases like Databricks Vector Search.
  
  Requirements
• 8–10+ years of ML Engineering experience.
• Strong background across classic ML, GenAI, and Agentic systems.
• Excellent English and Communication Skills (clear, concise, to the point). VERY Important.
• High Agency - High ownership. Takes initiatives able to drive outcomes independently.
• Founding member of the AI CoE, with significant influence over technology choices, frameworks, and engineering standards.

• STEM degree from a reputable university.

  Details:

• AI/ML Expertise: Significant hands-on experience deploying AI/ML solutions to production. While deep GenAI experience is preferred, strong traditional ML Engineering backgrounds is a big plus.
• Education: Bachelor's or Master's degree in Computer Science, Mathematics, Engineering, Physics, or a related field.
• Expert-Level GenAI Engineering: Deep hands-on experience building production RAG systems. You understand the nuances of chunking strategies, hybrid search (dense + sparse), re-ranking, and context management.
• Agentic Workflow Experience: You have built systems where LLMs use tools, maintain state, and execute multi-step plans. You are familiar with the challenges of loops, recursion, and hallucination mitigation.
• Strong Software Engineering Foundation: You are a software engineer first. You write clean, maintainable Python code and understand distributed systems, API design (FastAPI), and cloud architecture.

• Strategic Thinking: You can balance the "bleeding edge" with "boring technology" to deliver pragmatic business value. You understand the trade-offs between cost, speed, and accuracy.

   

  Responsibilities

• Design and build scalable, reusable AI services (e.g., the "Adaptive RAG Engine," "Text-to-SQL Service," and "Agentic Runtime") that product teams can easily integrate.
• Serve as the technical authority on Generative AI. Mentor senior engineers, conduct code reviews, and drive the adoption of engineering best practices (testing, CI/CD for AI, evaluation).
• Lead the development of multi-agent systems capable of complex reasoning, planning, tool use, and self-correction (using frameworks like LangGraph or Semantic Kernel).
• Load datasets and knowledge sources from external systems.
• Shape and transform data to optimize it for Hybrid or Semantic Search, providing the best possible knowledge context to LLMs.
• Move prototypes to production while focusing on latency, cost optimization, and reliability. Implement robust evaluation pipelines (LLM-as-a-Judge) to measure quality and prevent regression.
• Implement guardrails for PII protection, hallucination detection, and prompt injection defense.

• Partner with product teams to de-risk their AI roadmaps. Translate abstract business requirements into concrete architectural patterns.

   

  We offer

• US and EU projects based on advanced technologies.
• Competitive compensation based on skills and experience.
• Regular performance appraisals to support your growth.
• Flexibility in workspace, either remote or in one of our development offices.
• Comprehensive medical insurance, including dental and massages.
• Personalized learning program tailored to your interests and skill development.
• Sport reimbursement program for onsite and online activities.
• Bonuses for recommendations of new employees.
• Bonuses for article writing, public talks, and other activities.
• 20 vacation days, 10 national holidays and 5 sick leaves.
• Maternity leave policy and family days off.
• Free tech webinars and meetups organized by Svitla.
• Welcome and anniversary presents, gifts for children, and more.
• Regular corporate events and meetups.
• Awesome team, friendly and supportive community!