Erick Cobos is a Senior Data Engineer and AI practitioner with 11 years of experience building large-scale data pipelines, managing multi‑terabyte datasets, and producing ML models from statistical methods to generative transformers. He blends a strong software-engineering foundation with cloud-native production experience, having led dataset engineering and multi-node GPU training (up to 16 H100s) for text-conditional music generation and constructed a 45 TB music corpus. His research-driven background spans neuroscience and biomedical ML—contributing to calcium imaging tooling (CaImAn) and publications in Nature, NeurIPS, ICLR, and Nature Neuroscience—where he developed interpretable generative models and end-to-end imaging pipelines. Erick is pragmatic about reproducibility and openness, supporting scientific software and improving core algorithms for spike deconvolution and spatial–temporal integration. Now based in Tübingen, he focuses on turning complex experimental and multimodal data into deployable, impact-driven products.
11 years of coding experience
5 years of employment as a software developer
Master of Science - MS, Artificial Intelligence, Master of Science - MS, Artificial Intelligence at Tecnológico de Monterrey
Exchange Program, Computer Science, Exchange Program, Computer Science at EPFL
Exchange Program, Computer Science, Exchange Program, Computer Science at Texas A&M University
Computational toolbox for large scale Calcium Imaging Analysis, including movie handling, motion correction, source extraction, spike deconvolution and result visualization.
Role in this project:
Back-end Developer & Data Scientist
Contributions:40 commits, 6 PRs, 2 comments in 3 months
Contributions summary:Erick significantly contributed to the `caiman` repository, a computational toolbox for calcium imaging analysis. Their work involved modifying core functions related to spatial component updates, including the handling of empty background components, indicating a focus on optimizing the core algorithms. The user also refined the integration of temporal and spatial components, adjusting parameters to facilitate AR process handling and spike inference, demonstrating a focus on data analysis and model optimization within the calcium imaging domain. Further contributions included a focus on deconvolution methods.
Visualize features cells are responsive to via gradient ascent.
Contributions:2 reviews, 20 commits, 16 pushes in 10 months
ascentreactvisualizegradient-ascentgradient
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.