Charlie Heus
Software Engineer at Third Culture Strategies
New York, New York, United States
Summary
👤
Senior🎓
Top SchoolCharlie Heus is a software engineer based in New York with 11 years of hands-on experience and a BA in Computer Science from Columbia University. He blends research-driven thinking with practical engineering, having worked on LLM evaluation for low-resource translation and contributed backend motion-correction and spike-sorting improvements to the widely used SpikeInterface Python library. His background spans web performance and product-focused internships—where he halved iOS load times and boosted SEO engagement—to biomedical data pipelines that categorized millions of hypothalamic cells for diabetes research. Comfortable across Python, R, and SQL, he builds evaluation pipelines, automated ETL, and algorithmic features that move experimental ideas into production. Colleagues describe him as methodical and curious, often surfacing subtle algorithmic improvements (e.g., robust LSQR updates and non-rigid registration refinements) that yield outsized impact. He is actively seeking a full-time engineering role where he can combine algorithmic research and scalable implementation.
11 years of coding experience
2 years of employment as a software developer
Bachelor of Arts - BA Computer Science, Bachelor of Arts - BA Computer Science at Columbia UniversityThe Churchill School and Center
Github Skills (15)
algorithm10
data-structures10
algorithms10
machine-learning10
python10
data-structure10
numpy10
data-science9
pytorch9
electrophysiology8
scipy8
spike8
neuroscience8
sorting8
numba7
Programming languages (8)
Github contributions (5)
SpikeInterface/spikeinterface
Jun 2022 - Jun 2022
A Python-based module for creating flexible and robust spike sorting pipelines.
Role in this project:
Back-end Developer / Data Scientist
Back-end Developer / Data ScientistContributions:53 reviews, 25 commits, 17 PRs in 14 days
Contributions summary:Charlie primarily contributed to the `spikeinterface` library by implementing and refining motion correction and spike sorting algorithms. Their work included enhancing the `motion_estimation` module with functionalities like normalized cross-correlation, robust LSQR updates, and integration of different convolution engines (NumPy and Torch). They also added options to upsample motion data and incorporated improvements to the non-rigid registration process. Additionally, they introduced enhancements and bug fixes for the ISO-CUT algorithm.
pythonspike-sortingneuroscienceelectrophysiology
cwindolf/ffn
Jan 2019 - Aug 2020
A fork of google/ffn for asynchronous distributed training on SLURM clusters
Contributions:387 commits, 10 PRs, 268 pushes in 1 year 7 months
asynchronousdistributed-trainingslurm
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.