Brandon Duderstadt
Founder & CEO at Wikimedia New York City
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolBrandon Duderstadt is a founder, CEO, and machine learning engineer with a decade of experience building high-impact AI products and companies from research prototypes to revenue-generating platforms. He led Nomic AI from whiteboard to over $17M raised and $1M+ in revenue, and contributed core engineering to widely used open-source projects like GPT4All and Nomic (including an AtlasClient for large-scale neural database workflows). His work blends research and product: publications at top conferences, large-scale visualization work (world’s largest T-SNE), and production ML wins such as halving model memory and slashing error rates in deployed systems. Currently he runs Calcifer Computing, explores computational semiotics as Johns Hopkins faculty, and serves on the Wikimedia NYC board and as an advisor to Nomic AI. Known for an interdisciplinary streak—describing his work as “eudaemonic instrument manufacturing”—he combines rigorous applied ML with creative, instrument-like systems that probe cognition and meaning.
10 years of coding experience
12 years of employment as a software developer
Johns Hopkins University
Github Skills (31)
data-visualizations10
webgl10
transformers10
pytorch10
javascript10
python10
machine-learning10
data-visualisation10
llm-inference10
plaintext10
data-visualization10
embedding10
clustering10
api-design9
evaluation9
Programming languages (6)
Github contributions (5)
nomic-ai/nomic
Aug 2022 - Jan 2023
Interact, analyze and structure massive text, image, embedding, audio and video datasets
Role in this project:
ML Engineer
ML EngineerContributions:73 reviews, 32 commits, 35 PRs in 5 months
Contributions summary:Brandon significantly contributed to the `nomic-ai/nomic` repository by implementing and refining the `AtlasClient` class, a core component for interacting with Nomic's neural database. Their work included creating and updating text and embedding projects, adding text and embeddings, and integrating progressive data updates. Furthermore, the user addressed error handling within data upload processes, and included the ability for users to specify projection parameters.
pythonmachine-learningatlas-databaseatlascosine-similarity
nomic-ai/gpt4all
Mar 2023 - Jul 2023
GPT4All: Run Local LLMs on Any Device. Open-source and available for commercial use.
Role in this project:ML Engineer
ML EngineerContributions:1 review, 6 PRs, 8 pushes in 4 months
Contributions summary:Brandon primarily focused on developing and implementing evaluation scripts for machine learning models within the GPT4All project. They created scripts to evaluate model perplexity and generate continuations based on prompts. The contributions involved setting up model configurations, reading evaluation data, and computing metrics. The user also worked on generating and visualizing perplexity histograms for different models.
ai-chatllm-inference
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.