Matthew Scholefield
Software Engineer at Databricks
San Francisco, California, Taiwan
Summary
🤩
Rockstar🎓
Top SchoolMatthew Scholefield is a seasoned software engineer based in San Francisco with 31 years of experience building robust back-end systems, ML features, and CI/CD pipelines. Currently at Databricks, he brings hands-on expertise from voice assistant projects at Mycroft AI where he redesigned core audio listeners, implemented wake-word RNNs, and deployed end-to-end ML and web stacks. He pairs practical systems thinking — from Kafka-powered metrics pipelines and dockerized microservices to S3-backed artifact publishing — with a minimalist design ethos captured by his GitHub motto “Simplicity is elegance.” Notably, his contributions to widely used open-source Mycroft components improved real-time audio robustness and cross-version compatibility while automating release workflows. He holds a CS degree from UIUC and has a track record of reducing latency and operational complexity in production voice and test infrastructures.
30 years of coding experience
3 years of employment as a software developer
University of Illinois Urbana-Champaign
Stackoverflow
3,106reputation
1.2mreached
24answers
42questions
python
top-5%
Github Skills (16)
speech-recognition10
audio-processing10
nlp10
python10
natural-language-processing10
build-automation10
cicd10
bash9
tensorflow8
sys7
embedded7
asynchronous6
floating-point6
fme6
rounding6
Programming languages (31)
Github contributions (5)
MycroftAI/mycroft-precise
Oct 2017 - Aug 2020
A lightweight, simple-to-use, RNN wake word listener
Role in this project:
Full-stack Developer & DevOps Engineer
Full-stack Developer & DevOps EngineerContributions:2 releases, 1 review, 237 commits in 2 years 9 months
Contributions summary:Matthew's commits primarily focus on improving the compatibility and functionality of the wake word listener. They increased compatibility across different Python versions by modifying the codebase. The user also added a setup script and a publish script to automate the build and release process, setting up the CI/CD pipeline. They also implemented a system for generating and uploading artifacts to S3.
raspberry-pivoice-controlvoice-recognitionrnnwake
MycroftAI/mycroft-core
Jun 2016 - May 2019
Mycroft Core, the Mycroft Artificial Intelligence platform.
Role in this project:Back-end Developer
Back-end DeveloperContributions:5 releases, 280 commits, 244 PRs in 2 years 11 months
Contributions summary:Matthew focused on refactoring and improving the audio listening component of the Mycroft AI platform. They rewrote the listener, including the mic and speech recognizer logic, incorporating features like a wake word recognizer and phrase recording capabilities. Code modifications include adjusting audio threshold levels and adding features like timeouts to improve the robustness of the audio processing within the system. The user also made fixes to the try_wake_up method within the listener.
pythonnatural-language-processingraspberry-piartificialnatural-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.