David Przybilla is a pragmatic software engineer with 13 years of experience building backend, infra and ML-enabled systems, currently advancing blockchain and data tooling at Chainlink Labs from Tokyo. He has led ML teams at Optimizely, productionized recommendation systems and scalable A/B pipelines handling millions of interactions, and built reproducible ML infrastructure and DataOps platforms in both startup and enterprise contexts. A polyglot in Python and Go with deep NLP roots (MS in Computational Linguistics), he has contributed to notable open-source projects such as DBpedia Spotlight and added Spanish support to the Natural library. David blends hands-on engineering—real-time ingestion at HENNGE and infrastructure-as-code guidance—with mentoring and cross-team integration, and maintains an active interest in speeding ML lifecycles and reproducible data science.
14 years of coding experience
12 years of employment as a software developer
Software Engineer, Computer Software Engineering, Software Engineer, Computer Software Engineering at Universidad del Valle (CO)
Master of Science (MS), Computer Linguistics, Master of Science (MS), Computer Linguistics at Universität des Saarlandes
DBpedia Spotlight is a tool for automatically annotating mentions of DBpedia resources in text.
Role in this project:
Backend Developer & QA Engineer
Contributions:17 commits, 3 PRs, 1 push in 1 year 2 months
Contributions summary:David primarily focused on improving the core functionality of the DBpedia Spotlight project, specifically related to stemming and disambiguation. They added and fixed tests for the SnowballStemmer, ensuring its correct implementation. Additionally, the user made several changes to the core disambiguation logic, particularly in the `SpotlightModel` and `DBTwoStepDisambiguator` classes, demonstrating a focus on the backend logic of the project. The user also made a fix for a bug related to space in FSA tokenizer, and added and refactored the FSA testing.
Contributions summary:David primarily contributed to adding Spanish language support for the `natural` natural language processing library. Their work involved adding a stemmer, stop words, and a tokenizer specifically for the Spanish language. This included implementing the core stemming logic, providing a list of stop words, and creating a tokenizer tailored for Spanish text. The contributions extended the library's capabilities for natural language processing tasks in Spanish.
nlpkebab-casecorenlpfacilitiesnatural-language
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.