Benito Geordie

Benito Geordie is an AI engineer at ThirdAI. Benito previously worked as a research intern at Carnegie Mellon University and as an undergraduate computer science researcher at Rice University.

At Carnegie Mellon University, Benito contributed to an open-source high-performance conflict-free replicated data type (CRDT) library. Benito also wrote an API for composing CRDTs that are not mutually commutative for building complex collaborative abstract data types. As a demonstration of a practical use case, they built an intention-preserving collaborative rich text editor using this API.

At Rice University, Benito implemented scalable data structures with locality sensitive hashing (MinHash, L2Hash, signed random projections). Benito also first authored a paper on using RACE, a Bloom filter-based data structure, for diversity sampling in metagenomic studies in a streaming setting. Their work achieved the highest species diversity compared to composable coresets, Diginorm, and random sampling, while consuming up to 30,000x less RAM than Diginorm, and running at up to 200x the throughput of composable coresets. These results were submitted to ISMB 2020.

Benito Geordie completed their Bachelor of Science at Rice University in the field of Computer Science. Benito then went on to study Higher Level Physics, Mathematics and Economics at Sekolah Pelita Harapan.

They are on a team with Shashank Sonkar - AI Engineer, David Torres - AI Engineer, and Minghao Yan - AI Engineer. Their manager is Tharun Medini, CTO & Co-Founder.