Senior ML Engineer (Knowledge Graphs)
First seen by Alion on Oct 16, 2026.
About Osavul
Osavul is a product company building an AI-driven intelligence platform that helps governments, international organisations, enterprises, and NGOs detect and counter hybrid threats.
We work with large-scale, multilingual data from social media, news, geospatial, and other open sources, helping analysts understand actors, events, narratives, and risks across sources, locations, and time.
Role Overview
We’re looking for a Senior ML Engineer to work on the shared AI layer behind our intelligence products, with a focus on NLP, information extraction, and knowledge graphs.
You’ll work on turning unstructured data into structured knowledge: extracting entities and relationships, resolving them across sources, and connecting them through knowledge graphs.
The role combines ML/NLP work with production engineering. You’ll choose and evaluate the right approaches, integrate models into large-scale data pipelines, and own solutions from experimentation and technical design through to production.
Requirements
5+ years of professional software engineering experience.
Upper-intermediate (B2) English or higher.
Engineering Skills
Strong Python skills and experience building and maintaining production ML services.
Experience building asynchronous, high-throughput services, including concurrent model and API calls, retries, rate limits, and failure handling.
Hands-on experience with FastAPI and Docker.
Experience integrating ML components into distributed batch or streaming pipelines using Celery, RabbitMQ, Apache Kafka, or similar technologies.
Good understanding of databases, including PostgreSQL, Elasticsearch, and vector databases.
Strong software engineering practices around testing, debugging, maintainability, and production support.
AI Skills
3+ years of hands-on experience in ML/NLP engineering, including taking models to production.
Production experience with information extraction from unstructured text, including entity and relation extraction, entity resolution, and entity linking.
Hands-on experience designing and building knowledge graphs, including ontology / graph schema design and graph databases such as Neo4j. GraphRAG experience is relevant, but should be complemented by broader knowledge graph experience.
Strong understanding of traditional NLP, transformer architectures, and language models, with the ability to choose the right approach based on the problem and production constraints.
Experience with deep learning frameworks such as PyTorch.
Experience evaluating ML/NLP systems, defining relevant quality metrics, and monitoring performance in production.
Nice to Have
Experience fine-tuning or self-hosting open-source models using approaches such as LoRA / PEFT, SFT, vLLM, or similar tools.
Experience with multimodal models, computer vision, or geospatial data.
Background in threat intelligence, OSINT, geospatial intelligence, or defence.
We Offer
Real ownership over the knowledge graph direction and the opportunity to bring ideas from experimentation into production.
Remote-first setup across Ukraine and Europe.
Unlimited paid time off and company-wide holidays.
Coverage for software, AI tools, and coworking.
Annual budget for conferences, courses, and professional development.
Hiring Process
Recruiter interview → ML interview with a short live coding session → engineering interview → final interview with leadership → reference check and offer.