Skip to content
View BibeshT-TXST's full-sized avatar
✌️
Hello there!
✌️
Hello there!

Block or report BibeshT-TXST

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
BibeshT-TXST/README.md

Typing SVG


Computer Science Junior at Texas State University · Student Worker @ UL Syetems
I build systems at work and train models at home. Currently deep in secure inference and (somehow) quantum ML.


LinkedIn Blog Email

What I do

At work, I'm part of the Texas State University Libraries Systems Team. I build full-stack features, write unit and integration tests, set up CI/CD pipelines to university RHEL servers, and add observability metrics in a real production environment.

Outside of work, I focus on applied ML and health tech. Right now I'm drawn to secure inference i.e. how clinical AI can be practical and trustworthy at the same time.

2026 Activity

Commits Projects


GitHub Streak


Activity Graph

Projects

🔬 SightX  ·  Feb – Apr 2026  ·  Repo  ·  Blog

A clinical AI screening tool for diabetic retinopathy: A leading cause of preventable blindness that often shows no symptoms until too late.

  • Trained ResNet-50 via transfer learning to classify retinal fundus images across 5 severity grades
  • End-to-end microservice stack: React → Node.js → FastAPI → PostgreSQL, containerized with Docker
  • Built with a personal reason: diabetes runs in my family

PyTorch ResNet-50 FastAPI React Node.js PostgreSQL Docker Supabase

⚛️ PlateauX  ·  May 2026 – Present  ·  Repo

A toolkit for diagnosing and fixing barren plateaus in quantum ML circuits: The problem where quantum models train but stop improving.

  • Early-stage; actively building diagnostic utilities and mitigation strategies
  • Sits at the cross-section of quantum physics and classical ML optimization, genuinely uncharted territory for me

Python Quantum ML PennyLane

🏛️ UL Asset Management System  ·  Texas State University Libraries  (Internal)

Production asset management system used by the university libraries team. I'm the primary developer on V2.

  • Rebuilt frontend in Next.js (TypeScript); migrated from a Dockerized PostgreSQL container to the university's centralized DB
  • Deployed via GitHub Actions → RHEL production server; NGINX load balancer; Argon2 + pepper auth
  • V1 live and in use · V2 actively in development

TypeScript Next.js Node.js PostgreSQL Docker NGINX GitHub Actions

Stack

Programming Languages

Python TypeScript JavaScript C++ Java SQL

Frontend

React Next.js MUI shadcn/ui

Backend

Node.js FastAPI PostgreSQL

Machine Learning

PyTorch scikit-learn Jupyter

DevOps

Docker GitHub Actions NGINX Linux

Exploring

AWS Supabase Quantum ML

A bit more

  • Spent ~2 years as a research coach at the university. Helping students find what they need taught me that clear communication is its own kind of skill.
  • Former middle school math tutor. Engineering problems are usually easier (usually).
  • I go back to DSA and first principles often. Understanding code matters more than shipping fast — though both matter.
Atomic Commits > Atomic Habits :P

Pinned Loading

  1. SightX SightX Public

    AI-powered Diabetic Retinopathy detection using ResNet-50, FastAPI, React, Docker and PyTorch. Full stack clinical screening system with Bayesian decision theory and TTA ensemble.

    JavaScript

  2. PlateauX PlateauX Public

    PlateauX is a Barren Plateau Diagnosis & Mitigation Toolkit. This is an on going project where I dive into the land of quantum physics and much more.

    1