Roman Grebennikov
Principal Engineer at Delivery Hero
Berlin, Germany
Summary
🤩
Rockstar🎓
Top SchoolRoman Grebennikov is a Principal Engineer based in Berlin with 14 years building large-scale data processing, search and ML-heavy systems. He specializes in Learn-to-Rank, embeddings, dense retrieval and productionizing NLP/LLM-based relevance solutions, and has hands-on experience from feature prototyping to performance engineering. At Delivery Hero he improved food search with dense retrieval and fine-tuned SBERT/PyTorch models while driving technical strategy through RFCs, MVPs and team mentorship. He is the creator and maintainer of Metarank, an open-source low-code LTR engine with 2k+ GitHub stars, and contributes backend Scala code such as healthcheck APIs using http4s. Roman also founded Nixiesearch to simplify Elasticsearch with S3-backed storage and auto-scaling, and has a PhD-level academic background in computer science informing his pragmatic, functional-programming-influenced approach. Colleagues know him as a search nerd who can argue about embeddings and LTR for hours while still shipping reliable, production-grade systems.
14 years of coding experience
13 years of employment as a software developer
PhD, Computer Science, PhD, Computer Science at Voronezh State University
English, Russian, German
Stackoverflow
3,308reputation
262kreached
83answers
4questions
amazon-s3
top-5%
cassandra
top-5%
Github Skills (13)
back-end-development10
scala10
amazon-s39
cassandra9
dependency-management9
java6
datastax6
junit6
cql6
sbt6
recovery6
jar6
integration-testing6
Programming languages (12)
Github contributions (5)
metarank/metarank
Aug 2020 - Jan 2023
A low code Machine Learning personalized ranking service for articles, listings, search results, recommendations that boosts user engagement. A friendly Learn-to-Rank engine
Role in this project:
Back-end Developer
Back-end DeveloperContributions:45 releases, 10 reviews, 383 commits in 2 years 5 months
Contributions summary:Roman primarily focused on building the healthcheck API for the project. They implemented the API using Scala, using the http4s framework. In addition to API work, they were also responsible for integrating and managing the dependencies used in the project, as evidenced by changes to the `build.sbt` and `project/plugins.sbt` files.
watsondataminingfeature-extractiondata-sciencelow-code
metarank/lightgbm4j
Nov 2020 - Jan 2023
Java LightGBM binding
Contributions:20 releases, 57 commits, 87 PRs in 2 years 2 months
javalightgbmbinding
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.