We are seeking Lead AI/ML Engineer with complex visionary approach and proactive position to lead the development of advanced AI-driven systems for intelligent document understanding and automated data extraction. This includes building robust pipelines for parsing complex, multi-format documents as well as developing autonomous AI agents capable of navigating web platforms and collecting unstructured data at scale.
Key Responsibilities:
- Lead the design and implementation of a hybrid AI architecture combining traditional OCR pipelines with modern LLMs and multimodal models
- Architect intelligent agents capable of navigating web sources and autonomously collecting document data across multiple formats (PDF, Excel, Word, JSON, images, etc.)
- Drive innovation in document understanding through structured parsing (headings, sections, tables), multimodal layout analysis, and semantic search
- Guide the development of an agent-orchestrator that dynamically selects optimal pipelines and models based on document characteristics
- Oversee integration with vector databases and metadata-driven storage for fast, structured access and retrieval
- Enable the automatic generation of structured response templates (e.g., JSON) from unstructured data using LLMs
- Actively contribute to R&D strategy, Proof-of-Concept development, and technology scouting for emerging AI tools and techniques
- Collaborate cross-functionally with software engineers, DevOps, and product stakeholders to ensure scalable, production-ready solutions
- Mentor and grow the AI/ML team, fostering a culture of experimentation, ownership, and continuous learning
Must-Have Qualifications:
- 5+ years of experience in Machine Learning, Deep Learning, or Applied AI
- Strong Python skills and experience with ML libraries such as PyTorch, TensorFlow, Keras
- Deep understanding of LLMs, including fine-tuning, evaluation, and integration (e.g., using LangChain, LlamaIndex, Transformers, FAISS)
- Experience with AI agent development and prompt engineering
- Familiarity with document processing challenges: OCR, NLP/NLU, multimodal inputs.
- Solid understanding of vector databases and semantic data retrieval
- Ability to translate business needs into ML/AI solutions and communicate insights effectively
- Experience working in dynamic R&D environments with evolving priorities
- Upper Intermediate or higher level of English proficiency
Nice-to-Have:
- Experience with Computer Vision tasks and layout analysis
- Familiarity with LLMOps and orchestration of ML pipelines
- Knowledge of DevOps/MLOps practices: Docker, Kubernetes, CI/CD
- Experience with both SQL, NoSQL, and vector-based storage systems
- Prior work on intelligent agents for automated data collection/scraping
We offer:
- Flexible schedule
- Horizontal management and communication directly with project stakeholders
- 20 days of paid vacation
- Paid sick-leaves
- Annual activities budget — we endorse your wish to self-improvement
- Remote work