Mike Innes
England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolMike Innes is a software engineer and language/tooling expert with 13 years of experience building production-grade systems for machine learning and scientific computing. He led core work on Julia’s ML ecosystem—co-creating Flux and Zygote, influential autodiff and ML libraries cited thousands of times—and has driven novel probabilistic and query-engine work at RelationalAI that sped queries by orders of magnitude. Known for elegant, well-architected tools that bridge research and deployment, he combines deep PL/compilers skills (Julia core, JAX contributions) with hands-on ML engineering (neural ODEs, SciML). He writes and speaks frequently for both technical and general audiences, and his essays have attracted attention beyond the community. Based in Edinburgh, he enjoys unifying disparate approaches such as Bayesian inference and simulation to tackle hard, real-world problems.
13 years of coding experience
6 years of employment as a software developer
University of Oxford
Github Skills (68)
restructuredtext10
float3210
ode10
python10
deep-learning10
floating-point10
website-design10
backend10
shapes10
markdown10
statistics10
rs10
document-generation10
editor10
numpy10
Programming languages (20)
Github contributions (5)
FluxML/model-zoo
Jun 2017 - May 2020
Please do not feed the models
Role in this project:
ML Engineer
ML EngineerContributions:201 commits, 43 PRs, 164 pushes in 2 years 11 months
Contributions summary:Mike primarily focused on developing and modifying machine learning models within the `fluxml/model-zoo` repository. Their contributions involved debugging and refining existing models, specifically within the `phonemes` directory, including changes to model architectures and data processing pipelines. They implemented prediction functionality and tweaked model parameters for improved performance. The user also contributed examples of machine learning applications within the repository, with focus on model training and evaluation.
deep-learningmachine-learningfluxfeedjulia
FluxML/Zygote.jl
Apr 2018 - Jul 2020
21st century AD
Role in this project:Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:5 releases, 764 commits, 111 PRs in 2 years 3 months
Contributions summary:Mike was primarily focused on developing the Zygote.jl library, as indicated by file modifications and commit messages. They implemented core components related to automatic differentiation, including basic IR format, forward-mode implementation, and support for handling control flow. The contributions also involved addressing code errors, refactoring, and incorporating new functionality to enhance the library.
control-flowautomatic-differentiationmachine-learningjulia-compilerjulia
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.