Backend developer for AI evaluation platform. Developed RAG pipelines with LangChain & Pinecone Vector DB, Google OAuth, and quiz generation with Zod schemas. Over 150 commits across a 4-person team.
Aakash Bhagavathi
アーカーシュ・バガヴァティ
“Selected for UTD ACM Projects (~20% acceptance rate) — Co-built Socratic, awarded 2nd Best Design & 3rd Overall”
UT Dallas CS ('28) · ACM Backend Engineer · UTD SCC Software Lead · JLA Officer (N3 Japanese)
SELECTED ARTIFACTS
Predictive ML model built in Python with Scikit-Learn (Logistic Regression & Random Forest) to analyze lifestyle data, clean feature sets with Pandas, and forecast sleep disorder risks.
Nintendo Switch emulator port optimizations exploring Tegra X1 SoC capabilities, ARM64 architecture, memory bandwidth tuning, and custom compilation pipelines.
Led a team of 4 software engineers to rebuild the organization's web platform from scratch using React and Firebase, migrating off Wix for complete custom code control.
Programmed vehicle physics, tilemap track generation, and collision detection in GDScript for Godot Engine. Released on itch.io for UTD SGDA and Godot Wild Jam.
TECHNICAL COMPETENCY
Primary Mastery: Java (Data Structures & Algorithms, Custom Stacks, Sliding Window, Systems)
Proficient: Python, C++, C#, Swift, GDScript, JavaScript (React, Node)
Low-Level Systems: Assembly (ARM64 / x86), Metal API, Vulkan API, Tegra X1 Architecture
Languages: English (Native), Japanese (N3 Level Comprehension · ~1,500 Vocab Pool)