Alistair Muldal
Scientific Researcher at DeepMind
London, England, United Kingdom
Summary
🤩
RockstarAlistair Muldal is a research scientist at DeepMind with 13 years of software engineering experience focused on reliable, scalable scientific tooling and simulation infrastructure. He blends back-end development, test automation and DevOps—contributing to high-profile open-source projects such as dm_control and PyTables where he fixed memory leaks, optimized builds, added POSIX-style softlinks, and hardened CI for headless testing. His work on dm_control improved MuJoCo-based RL environments and testing frameworks, and he has strengthened plotting and QA in matplotlib through targeted test coverage. Comfortable across low-level performance fixes and higher-level automation, he brings a pragmatic eye for reproducibility and maintainability in research codebases. Based in London, he pairs DeepMind research priorities with hands-on contributions that keep large scientific software ecosystems robust and testable.
14 years of coding experience
Stackoverflow
73,098reputation
9.0mreached
691answers
45questions
image
top-1%
signal-processing
top-5%
matplotlib
top-1%
sorting
top-5%
animation
top-5%
concatenation
top-5%
Github Skills (70)
python10
multidimensional-arrays10
testing10
bash10
while-loop10
matrix10
statistics10
linux10
numpy10
performance-optimization10
data-visualization10
test-automation10
data-visualizations10
scipy10
cicd10
Programming languages (10)
Github contributions (5)
google-deepmind/dm_control
Jan 2018 - Nov 2022
Google DeepMind's software stack for physics-based simulation and Reinforcement Learning environments, using MuJoCo.
Role in this project:
Back-end Developer & Test Automation Engineer
Back-end Developer & Test Automation EngineerContributions:11 reviews, 159 commits, 18 PRs in 4 years 10 months
Contributions summary:Alistair's contributions focused on enhancing the `dm_control` codebase by refining existing methods, clarifying documentation, and introducing new testing functionalities. They improved the codebase through refactoring, such as renaming variables and modifying the APIs of methods like `action_spec` and `before_step`, as well as improving the testing framework. This included fixing a bug in the threaded decorator, and creating new testing cases within the existing Mujoco framework. Furthermore, improvements were made to incorporate the disabling of OpenGL rendering.
mujocoreinforcement-learning-environmentssimulationmachine-learningartificial-intelligence
google-deepmind/deepmind-research
Jan 2021 - Aug 2021
This repository contains implementations and illustrative code to accompany DeepMind publications
Role in this project:DevOps Engineer & Automation Engineer
DevOps Engineer & Automation EngineerContributions:18 commits, 1 PR, 20 pushes in 6 months
Contributions summary:Alistair's commits primarily focus on configuring and enabling Travis CI tests for various projects within the `deepmind-research` repository. They ensured that the projects' dependencies were up-to-date by updating the `pip` installation and installing required packages. Additionally, the user modified the `run.sh` scripts to accommodate testing on the Travis CI platform, including disabling plotting for headless environments and correcting relative file paths.
pytorchimplementationsdeep-learningneural-networksmachine-learning
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.