Hi, I'm
Caleb Galdston_
ECE @ UCSB · Controls · Optimization · ML
Grad student at UCSB studying controls and optimization. Before that, math and CS at UCSD. Currently looking for internships in systems engineering, controls, and ML.
About

I'm an M.S. student in Electrical and Computer Engineering at UC Santa Barbara, with an emphasis in Control, Communications, and Signal Processing. Before that, I earned a B.S. in Mathematics-Computer Science with a minor in Data Science from UC San Diego.
Alongside grad school, I work remotely as a District Data Specialist for Fitchburg Public Schools in Massachusetts, where I build automated Java and SQL data pipelines, certify state reporting for a 5,500-student district, and built an enrollment projection model used for capacity planning.
My projects span a wide range: applying deep learning to climate model emulation, building recommendation systems, training RL agents in gym environments, and generating symbolic music with transformers.
Outside of work, I spend my time powerlifting, running, skiing, reading, and obsessively following the Red Sox.
Projects
Deep Learning For Climate Model Emulation
ConvLSTM model that emulates CMIP6 climate simulations, cutting RMSE 57% vs. a CNN baseline
VolleyballReinforcementLearning
DQN agent for SlimeVolleyGym, benchmarked against PPO, A2C, and CEM
Unconditional Symbolic Music Generation
Transformer that generates symbolic music (MIDI) from scratch
Vermont Business Recommender System
Recommender system for Vermont businesses built on review data
Goodreads Recommendation
Two-tower book recommender on the UCSD Goodreads dataset, with pgvector search
MLB Pitch Sequencing
Modeling pitch sequencing patterns in statcast data
GitHub Activity
Contact
Open to interesting problems and new opportunities. Reach out anytime.