Harald Husum
Senior Machine Learning Engineer at Intelecy
Oslo, Norway
Summary
🤩
Rockstar🎓
Top SchoolHarald Husum is a Senior Machine Learning Engineer based in Oslo with a decade of experience building production-ready ML and data systems. He blends deep Python engineering—improving type safety and async runtime code—with practical probabilistic modeling, evidenced by contributions to tensorflow/probability and core libraries like Streamlit, Trio, and YAPF. At Intelecy he progressed from Data Scientist to senior engineering roles, shipping models and infrastructure that bridge research and product. Known for thoughtful code maintainability (type annotations, tests, refactors), he favors clear, reproducible tooling that moves projects forward. Outside work he brings a developer’s curiosity to open source, often tackling subtle correctness and usability issues rather than flashy features.
10 years of coding experience
5 years of employment as a software developer
Norwegian University of Science and Technology
Norwegian, Norwegian, English, German
Stackoverflow
1,139reputation
16kreached
5answers
11questions
Github Skills (41)
probabilistic-programming10
python10
triplestore10
async10
mypy10
tripledes10
triple10
formatter10
triplet10
typehinting10
type-annotations10
structured-concurrency10
tensorflow10
formatters10
async-await10
Programming languages (12)
Github contributions (5)
python-trio/trio
Nov 2022 - Jan 2023
Trio – a friendly Python library for async concurrency and I/O
Role in this project:
Back-end Developer
Back-end DeveloperContributions:9 reviews, 50 commits, 17 PRs in 2 months
Contributions summary:Harald primarily contributed to the internal structure and type annotations of the Trio library, specifically focusing on the core run-time aspects. Their work included adding and modifying type hints for improved code clarity and maintainability. They addressed mypy errors to ensure type safety and consistency throughout the codebase. These changes appear to contribute to the overall quality and reliability of the asynchronous concurrency library.
python-librarypythontrioasynchronousasync-await
streamlit/streamlit
Oct 2021 - Jan 2023
Streamlit — A faster way to build and share data apps.
Role in this project:Full-stack Developer
Full-stack DeveloperContributions:71 reviews, 40 commits, 57 PRs in 1 year 3 months
Contributions summary:Harald primarily contributed to the Streamlit library by adding and improving type annotations. Their work focused on annotating various functions and classes across multiple files, including `__init__.py`, `markdown.py`, and elements like `snow.py`, `layouts.py`, `json.py`, `empty.py`, `balloons.py`, `plotly_chart.py`, and `exception.py`. They also updated the `write` and `doc_string` files, as well as the `time_widgets` elements, enhancing the codebase's maintainability and type safety.
data-analysispythondata-sciencedeep-learningfastest
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.