Stevan Matovic
Verified Expert in Engineering
Machine Learning (ML) Engineer and Developer
Novi Sad, Vojvodina, Serbia
Toptal member since October 13, 2026
Stevan is an ML engineer with 7 years of experience building production AI systems across enterprise search, RAG, recommender systems, NLP, and multimodal AI. He specializes in taking ML products end to end, from experimentation and evaluation to scalable inference, deployment, and production integration. Stevan will be a great addition to any team.
Portfolio
Experience
- Machine Learning Operations (MLOps) - 7 years
- Artificial Intelligence (AI) - 7 years
- Machine Learning - 7 years
- Python - 7 years
- PyTorch - 7 years
- AI Inference - 7 years
- Large Language Models (LLMs) - 5 years
- RAG Systems - 4 years
Work Experience
ML Engineer
OneTrust
- Spearheaded the migration of ML and data engineering processes to Databricks, optimizing workflow efficiency.
- Implemented a text classification model deployed on customer clusters, meeting specific client requirements.
- Developed and managed CI/CD pipelines for model deployment, streamlining release cycles.
- Worked on a clustering model, contributing to the analytical capabilities of the platform.
- Established model monitoring mechanisms to ensure optimal performance and reliability.
ML Engineer
SmartCat
- Developed and deployed a recommender system in the sports betting industry.
- Scaled the solution to millions of users for the recommender system in the sports betting industry.
- Implemented real-time inference for the recommender system in the sports betting industry.
- Created an automatic retraining and redeployment pipeline for the recommender system in the sports betting industry.
- Developed and deployed collaborative filtering recommender systems for online casino games.
- Built an automatic retraining pipeline for collaborative filtering recommender systems.
ML Engineer
Vivify Ideas
- Developed a fashion recommender system using computer vision technology, enhancing personalized shopping experiences.
- Created a content-based image retrieval service using computer vision technology.
- Developed an NLP product that extracts metadata from business agreements.
Project History
Enterprise Search Reranking with Cohere Rerank v4
Based on these results, I implemented the production integration across the search service, model marketplace, and platform infrastructure. I built a reusable reranker service abstraction and Cohere implementation, integrated reranking into the enterprise search pipeline behind a feature flag, preserved the existing scoring strategy as a fallback, added model configuration and pricing metadata, and implemented integration tests and deployment configuration.
The result was a production-ready architecture that allowed semantic search results to be reranked by relevance while retaining safe fallback behavior.
Education
Master's Degree in Computer Science
Faculty of Technical Sciences - Novi Sad, Serbia
Bachelor's Degree in Computer Science
Faculty of Technical Sciences - Novi Sad, Serbia
Skills
Libraries/APIs
PyTorch, TensorFlow, Keras, SpaCy
Tools
Jenkins, Prefect, Kafka Streams, Git
Languages
Python, Scala, JavaScript
Platforms
Azure, Databricks, Apache Kafka, AWS IoT, Jupyter Notebook, Docker
Frameworks
Spark, Django
Storage
Elasticsearch, PostgreSQL
Other
Machine Learning, Artificial Intelligence (AI), RAG Systems, Machine Learning Operations (MLOps), AI Inference, LLM Integration, Retrieval-augmented Generation (RAG), Large Language Models (LLMs), AI Agents, Agentic AI, MLflow, Computer Science, Information Retrieval, Semantic Search
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring