Suvan Kasina
AI Engineer • Data Scientist • Software Developer
About Me
- Full Name:Suvan Kasina
- Phone:+1 (210) 391-4711
- Email:suvan.kasina8@gmail.com
- Location:San Antonio, TX
- Download My Resume (PDF)
Hello There!
I'm a Computer Science undergraduate at the University of Texas at San Antonio focused on AI, machine learning, and data engineering. I build reproducible ML pipelines, scalable web apps, and production-focused tools that support research and product goals.
Recent work includes an AI Futures internship at MITRE (multi-node GNN training cluster, distributed PyTorch), software engineering at the UTSA Data Analytics Center (data pipelines, scrapers, and research tooling), and a production internship at Sports Media Inc. where I integrated conversational AI and voicebots.
I'm actively seeking AI/ML internships and roles where I can contribute to model development, data infrastructure, and deployable systems. Check out my projects below and my GitHub for source code and demos.
Work Experience & Education
-
AI Futures Intern — MITRE
Jun 2025 – Aug 2025
Built a distributed multi-node GNN training cluster (Kubernetes, Podman, Docker, VirtualBox), designed a Master-Worker training architecture and improved GNN training throughput using torch.distributed and PyTorch Geometric.
-
Software Engineer — UTSA Data Analytics Center
Feb 2025 – Apr 2025
Built web scrapers and AI-powered data pipelines (BeautifulSoup, Pandas), designed relational schemas, optimized SQL queries, and applied PyTorch models for exploratory analysis supporting research publications.
-
Computer Science Intern — Sports Media Inc.
Jun 2024 – Oct 2024Integrated an AI chatbot with a Twilio-based voicebot and LLMs, improving customer satisfaction by ~30% and automating backend workflows and web platforms to improve performance.
-
Computer Science Fellow — Headstarter AI
Jul 2024 – Sep 2024Collaborated on AI projects and rapid prototyping with a cohort of students focused on real-world ML/AI solutions.
-
Driver & Team Captain — FTC Team 15024
May 2021 – May 2024Led development of a competition robot (Java, Android Studio, TensorFlow, OpenCV). Achieved Semi-Regional Finals and UIL State selection, and mentored teammates in Java & robotics systems.
-
Education
-
B.S. Computer Science — University of Texas at San Antonio (Data Science)
2024 – 2027 (Expected) -
Tom C. Clark High School
Graduated 2024 — 4.0 GPA
Selected Projects
FastCAV — Explainability for Vision Models
Implemented the FastCAV framework in PyTorch to compute Concept Activation Vectors for large vision models and automated pipelines using Hydra, timm backbones, and visualization of concept evolution.
Skills
PyTorch, timm, ViT, Hydra, Docker
FareHunter — Rideshare Cost Optimizer
Full-stack Next.js application that ranks optimal pickup points to reduce fares using live mapping and surge data, deployed on Vercel.
Skills
Next.js, React, TypeScript, Node.js, MapBox
Premier League Match Predictor
Data pipeline and Random Forest model trained on 2000+ matches and 150+ features; dataset published on Kaggle with users and downloads.
Skills
Python, Pandas, scikit-learn
Hacker News Notifier
Daily webscraper that curates top Hacker News posts and emails subscribers; uses BeautifulSoup and SQLite.
Skills
Python, BeautifulSoup, SQLite, Email Automation
StudyFuze — Student Resource App
Flask + SQL full-stack app for resource sharing with secure auth, role-based access, and RESTful endpoints supporting a test cohort of users.
Skills
Flask, SQL, JavaScript, HTML/CSS
FTC Robot — Team 15024
Full codebase for the competition robot with autonomous TensorFlow/OpenCV routines and Java control systems.
Skills
Java, Android Studio, TensorFlow, OpenCV
