Ross Taylor
Co-Founder & CEO at General Reasoning
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolRoss Taylor is a founder and AI research leader with a decade of experience building large language models and startup technology, currently co-founding and leading General Reasoning from London. Previously he was LLM lead at Meta AI, driving reasoning and agent research on LLaMA 2/3 and leading scientific-model efforts in FAIR, and he co-created Papers with Code. His background spans founding Atlas ML (acquired by Facebook), quantitative analysis, and academic training at Cambridge and Warwick, blending rigorous MPhil-level research with product leadership. An active open-source contributor, he has extended Python time-series tooling (pyflux) with ARMA/GAS capabilities, reflecting a comfort across both statistical modeling and production ML systems. Colleagues know him for turning cutting-edge research into usable tools and teams that ship.
10 years of coding experience
8 years of employment as a software developer
Ipswich SchoolMaster of Philosophy (MPhil), Master of Philosophy (MPhil) at University of CambridgeBachelor of Science (BSc), Bachelor of Science (BSc) at University of Warwick
Github Skills (4)
time-series10
statistics10
python10
machine-learning9
Programming languages (5)
Github contributions (5)
RJT1990/pyflux
Feb 2016 - Dec 2018
Open source time series library for Python
Role in this project:
Data Scientist
Data ScientistContributions:112 commits, 8 PRs, 94 pushes in 2 years 10 months
Contributions summary:Ross primarily contributed to the implementation of new GAS/ARMA functions within the `pyflux` library. Their work involved creating and integrating methods for time series analysis, specifically focusing on the ARMA (Autoregressive Moving Average) class within the package. The commits show the addition of key code within the `arma.py` and `gas.py` files, indicating a focus on expanding the library's statistical capabilities for time series modeling and forecasting.
time-seriespythontimeseriestime-series-analysisstatistics
paperswithcode/torchbench
Jul 2019 - Oct 2019
Easily benchmark machine learning models in PyTorch
Contributions:67 commits, 14 PRs, 43 pushes in 2 months
pytorchdeep-learningmachine-learningmachine-learning-benchmarkbenchmark
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.