Brain–Computer Interfaces · Machine Learning · AgTech · Full-Stack Software
I build systems that turn noisy signals into meaning — from brain–computer interfaces to in-field nutrient sensors, and the ML, firmware, and cloud that run them.
About
Caleb works at the seam of neuroscience, machine learning, and software engineering. For most of a decade he's been decoding brain activity — fNIRS, EEG, and voltage imaging — and building the real-time systems, pipelines, and models that turn those signals into something usable.
His research runs from brain–computer interfaces and neurofeedback to explainable AI; his engineering runs from embedded firmware to ETL at scale to AI products in production. He's as interested in why a model makes a prediction as whether it's right. Today that means building Page Technologies' field nutrient sensors — the same decode-the-signal problem, now in soil and water instead of cortex.
Brain–Computer Interfaces
Real-time acquisition and decoding of fNIRS and EEG into discrete brain states.
Neurofeedback
Closed-loop games and interfaces driven by live signal classification.
Machine & Deep Learning
Supervised, unsupervised, and reinforcement learning across signal, image, and text.
Explainable AI
Interpreting predictions down to the contributing channel with DeepSHAP.
Signal & Image Pipelines
Motion correction, source extraction, and ETL for large microscopy and physiology datasets.
Production Systems
Embedded firmware in C with MQTT / AWS IoT, plus cloud APIs and interfaces.
AgTech · Page Technologies
At Page Technologies, Caleb builds the Autosampler — a field instrument that automatically samples, measures, and uploads nutrient levels (nitrate, potassium, calcium, ammonium) from greenhouses, soil, and water systems. It's the same problem as the rest of his work in a new medium: take a noisy analog signal and turn it into something you can act on. He owns it end to end — the firmware on the device, the models that read the chemistry, and the platform growers log into.
Firmware
Embedded · C / ESP32
Machine Learning
Sensor calibration · signal
Full-Stack
Cloud + web
Experience
Publications
37th Conference on Neural Information Processing Systems (NeurIPS) — Datasets & Benchmarks Track, 2023
@inproceedings{pyarelal2023tomcat,
title = {The ToMCAT Dataset},
author = {Pyarelal, Adarsh and Duong, Eric and Shibu, Caleb Jones
and Soares, Paulo and Boyd, Savannah and Khosla, Payal
and Pfeifer, Valeria and Zhang, Diheng and Andrews, Eric S.
and Champlin, Rick and Raymond, Vincent Paul
and Krishnaswamy, Meghavarshini and Morrison, Clayton
and Butler, Emily and Barnard, Kobus},
booktitle = {Thirty-seventh Conference on Neural Information Processing
Systems Datasets and Benchmarks Track},
year = {2023},
url = {https://openreview.net/forum?id=ZJWQfgXQb6}
}
M.S. Thesis, The University of Arizona, 2023
@mastersthesis{shibu2023decoding,
title = {Decoding Emotional Responses: A Comparative Study
of fNIRS and EEG Neuroimaging Techniques},
author = {Shibu, Caleb Jones},
school = {The University of Arizona},
year = {2023},
url = {https://repository.arizona.edu/handle/10150/670846}
}
Frontiers in Human Neuroscience — Brain-Computer Interfaces, 2022
@article{shibu2022explainable,
title = {Explainable artificial intelligence model to predict
brain states from fNIRS signals},
author = {Shibu, Caleb Jones and Sreedharan, Sujesh
and Arun, K. M. and Kesavadas, Chandrasekharan
and Sitaram, Ranganatha},
journal = {Frontiers in Human Neuroscience},
year = {2022},
doi = {10.3389/fnhum.2022.1029784}
}
5th International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS), IEEE, 2020, pp. 152–157
@inproceedings{shibu2020comparison,
title = {Comparison of classification performance of handpicked,
handcrafted, and automated-features for fNIRS-BCI system},
author = {Shibu, Caleb Jones and Sreedharan, Sujesh
and Arun, K. M. and Kesavadas, Chandrasekharan},
booktitle = {2020 5th International Conference on Intelligent
Informatics and Biomedical Sciences (ICIIBMS)},
pages = {152--157},
year = {2020},
publisher = {IEEE},
doi = {10.1109/ICIIBMS50712.2020.9336392}
}
Contact
Building something at the edge of brains and machines? Let's talk.