Sravanth Chowdary Potluri

Resume

About

Currently a Masters student at the University of Virginia, keeping up with the expanding horizons of computer science. I enjoy solving hard problems at the intersection of systems and intelligence with my skills.

Contact me at sravanth.chowdary.potluri@gmail.com

GitHub | LinkedIn

Education

University of Virginia

Master of Computer Science (MCS) (CGPA: 3.87/4) | Aug 2024-Dec 2025
  • Fall ’24 — Geometry of Data, Machine Learning in Image Analysis, NLP, Autonomous Mobile Robots
  • Spring ’25 — Software Analysis and Applications, Machine Learning in Graphs, Neural Networks
  • Fall ’25 — Convex Optimization for Engineering and Data Science, Signal Processing, ML and Control
  • Teaching Assistant for CS 2120: Discrete Math & Theory I — Fall ’24 and Fall ’25

Indian Institute of Information Technology, Design and Manufacturing, Kancheepuram

Bachelor of Technology in Computer Science and Engineering (Honors) — Distinction(CGPA: 9.19/10 ~ 3.9/4) | Aug 2020–May 2024

Experience

Software Development Engineering Intern — Amazon Web Services (AWS)

May 2025–Aug 2025
  • Worked on the EMR Serverless team to automate processing of limit‑increase requests and their workflows
  • Designed and built a distributed system using AWS and internal services to automatically process requests, reducing operational burden across 30+ AWS regions
  • The system, capable of processing thousands of requests per day, reduces average ticket processing time from 31 days to a couple of minutes

Research Assistant — Data Driven Streets (UVA Darden)

Oct 2024–Dec 2025
  • Working as an RA through UVA Darden on data and infrastructure
  • Building and maintaining data pipelines using AWS that scrape, process, and add thousands of rows of information to databases on traffic and pedestrian patterns in the D.C. area
  • Built a system using LLMs and in‑context learning to extract entities from emergency call transcripts, and a web app to display transcripts along an entity timeline

Machine Learning Research Intern — Ericsson

Jul 2023–Dec 2023
  • Proposed and developed a system using brain‑inspired neural networks (spiking, LMU RNN) to detect accidents in real time and communicate efficiently via predicted network nodes
  • Contributed to a system with graph spiking neural networks (GSNNs) for data‑rate prediction to route vehicles efficiently in optimal network scenarios
  • Proposed and built an ontology‑mapping architecture using semantic mapping with RoBERTa and GPT, enhanced via causal relationships
  • Programmed machine‑learning models and performed data analysis using Python and related libraries
  • Conference paper published and patent filed

Publications & Patents

Technical Projects

Hate Speech Detection Using Multi‑LLM Architecture

  • Classification using ensemble of fine-tuned LLMs with in-context learning
  • Two-tiered multi-LLM architecture using LLaMA models (1B, 3B)
  • 18% faster processing and 54% lower memory usage
  • Fine-tuning with Torchtune and ICL
  • Proposed: model calibration, LoRA/QLoRA for efficient fine-tuning

Goal‑Oriented Image Quality Assessment Using CNN

  • Novel image quality assessment powered by deep learning
  • Task-specific image quality evaluation
  • ResNet-based depth estimation on NYU-Depth-v2
  • Pearson correlation 0.89 and ~1900% efficiency improvement

Waze-Based Traffic Forecasting Using Graph Transformers

  • City-scale traffic prediction using STGformer (graph transformer)
  • Trained on Washington D.C. Waze data (4 years, ~20M jams)
  • Engineered road network graph; sparse subgraph sampling in PyTorch
  • Achieved R^2 ≈ 0.86

LLM vs LLM: An Approach to Fault Localization

  • Adversarial LLM framework for software fault localization
  • Gemma-3-12B generates buggy C++; a second model localizes faults
  • Custom fault injector for 5,000 C++ programs
  • QLoRA fine-tuned debugger improves accuracy from ~20–30% to ~55–65%
  • Demonstrates PEFT feasibility

Single‑Lead to 12‑Lead ECG Conversion Using RNN

  • LSTM network predicts multiple ECG leads
  • Predicts remaining 11 leads from a single lead
  • Noise removal: Butterworth filter and DWT (db-6 wavelet)
  • Experiments with noise removal strategies
  • Peak R^2 0.9 (some leads); peak average R^2 0.56

Incognito Text

  • Simple, minimalistic, anonymous texting web application
  • Chat rooms and anonymous accounts
  • Flask backend with relational database
  • Deployed on Heroku and AWS (PostgreSQL, SQLite)

Maze Solver Using AI

  • Heuristic and brute-force based maze solver
  • Pygame GUI with random maze generation
  • Solvers: BFS, DFS, A* (Manhattan/Euclidean)
  • Execution time analysis across 30 random mazes
  • A* mean 0.01s; BFS 0.029s; DFS 0.06s

Relevant Coursework

Technical Skills & Interests

Achievements & Awards

Contact

✉️ sravanth.chowdary.potluri@gmail.com

💻 GitHub | LinkedIn