Hi! I'm Chetan.
I'm an engineer finishing my M.S. in EECS at UC Berkeley — graduating August 2026. I've built things across the stack: full-stack apps, ML pipelines, a compiler from scratch, and a hardware piece that listens to a room and plays it back. Lately I've been most interested in how AI tools can make programming feel more intuitive, which is what my current research is about.
I'll be available for full-time software engineering roles from late August. Feel free to reach out at cgoenka@berkeley.edu.
Skills
Experience
Graduate Research Engineer · UC Berkeley EECS
Jun 2024 - PresentBuilt a code explanation system on top of CodeMirror and OpenAI API — line-level, block-level, and data-flow explanations rendered as interactive overlays right inside the editor. Wrote 700+ lines of custom editor extensions to make it work, iterated on it with user studies, and shipped a function-level summarization tool on the back of that feedback.
Course Staff, Data 100 · UC Berkeley
Aug 2023 - May 2024Helped 25+ students a week debug their Python and SQL for Data 100 — pandas, NumPy, scikit-learn, matplotlib — and graded 5,000+ assignments across two semesters, writing scripts to handle the scale and giving structured feedback on data pipelines and statistical implementations.
Data Science & Engineering Intern · Oracle
Jun - Aug 2022Automated the sales prospecting process: built an ML scoring pipeline in Python and SQL that ranked 500+ enterprise prospects across four products, replacing manual prioritization. Also built the ETL layer that brought in marketing campaign data and calculated attribution rates, and rewrote the SQL powering the exec dashboards into something actually efficient.
Featured Work
Projects
Python Compiler
A compiler I built from scratch in Java for a Python dialect — the whole journey from raw source through lexing, parsing, and type checking down to RISC-V assembly, plus a little web IDE that lets you watch each stage happen.
Java · Python · RISC-V (Assembly Language) · JFlex (Compiler Tool) · CUP (Compiler Tool)
Research
Code Explanations: Automated Hierarchical Descriptions of Program Behavior
A tool that layers AI-generated explanations right onto your code — per line, per block, and per parameter — so a program becomes easier to read and safer to change. Built on CodeMirror and GPT-4o, with interactive insights rendered directly in the editor.
Memory Bottle
An interactive bottle that captures a moment — its sound and its color — and pours it back out when you tip it over, filling the room with the audio and light of when you saved it. Underneath, an ESP32 and a handful of sensors talk to a small server that keeps the audio and lights in sync.