Ogundepo Odunayo
Applied Scientist at Masakhane
Kurashiki, Okayama Prefecture, Canada
Summary
🤩
Rockstar🎓
Top SchoolOgundepo Odunayo is an Applied Scientist with seven years of ML engineering and research experience, currently at Amazon after driving large-scale retrieval and LLM alignment work at LexisNexis. He holds an MMath in Computer Science from the University of Waterloo and has published at top AI venues while focusing on language modelling, NLP, and information retrieval for low-resource and African languages. His open-source contributions to flagship IR toolkits like Pyserini and Anserini show deep hands-on expertise in dataset integration, indexing, and test automation for multilingual corpora such as HC4, AfriBERTa and NeuCLIR. He has productionized bi-encoder entity resolution serving millions of daily requests, built automated annotation pipelines, and led AfroBench and AfriQA efforts—demonstrating a rare blend of research rigor and operational impact. Colocated between Kurashiki and Canada, he combines academic depth with practical system-building that improves retrieval and generation at scale.
7 years of coding experience
1 year of employment as a software developer
Bachelor of Science (B.Sc.) Petroleum Engineering, Bachelor of Science (B.Sc.) Petroleum Engineering at University of IbadanMMath Computer Science Computer Science, MMath Computer Science Computer Science at University of Waterloo
Yoruba, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (12)
lucene10
javas10
json10
information-retrieval10
python10
java10
testing10
data-science9
test-automation9
unit-testing9
data-structures8
data-structure8
Programming languages (9)
Github contributions (5)
castorini/anserini
Oct 2021 - Oct 2022
Anserini is a Lucene toolkit for reproducible information retrieval research
Role in this project:
Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:35 reviews, 28 commits, 54 PRs in 1 year
Contributions summary:Ogundepo contributed significantly to the Anserini project, focusing on enhancing its capabilities for information retrieval research, particularly within the context of the AfriBERTa and NeuCLIR corpora. The user implemented support for indexing and processing specific datasets, including a new collection type for AfriBERTa. They also added support for NeuCLIR. These changes demonstrate a focus on expanding the toolkit's data ingestion capabilities to support different languages and data formats.
reproducibleretrievalluceneinformation-retrievaljava
castorini/pyserini
Jul 2022 - Dec 2022
Pyserini is a Python toolkit for reproducible information retrieval research with sparse and dense representations.
Role in this project:Back-end Developer & Test Automation Engineer
Back-end Developer & Test Automation EngineerContributions:23 reviews, 8 commits, 25 PRs in 5 months
Contributions summary:Ogundepo primarily focused on adding and updating functionality related to HC4 datasets within the Pyserini toolkit. This involved implementing bindings, pre-built indexes, and test cases for different language versions of the HC4 dataset. Furthermore, the user refactored naming conventions and updated regression tests, demonstrating a commitment to code quality and maintainability. Their contributions directly supported the integration and testing of information retrieval capabilities with the HC4 dataset, showcasing proficiency in the project's core domain.
record-linkagepythonrepresentationssparsesimilarity
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.