Skip to content
View a6nkay-source's full-sized avatar
🏠
Working from home
🏠
Working from home

Block or report a6nkay-source

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.

Content in all repositories owned by your account will be closed.
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
a6nkay-source/README.md

Anshuman Kumar. I build and validate machine-learning systems for health, finance and the environment, then test them on data they have never seen. Interested in biology, business, technology and the future of AI.

Med-tech research

Voice-biomarker models for Parkinson's disease, validated across independent patient cohorts with calibration, domain-shift and leakage analysis. Surgical video segmentation in SurgiVision and on-device wellness signals in Anchor.

Business and finance

Portfolio research for the Wharton Global Youth investment competition: walk-forward models on 74 large-cap stocks, backtested from 2006 to 2026 after costs and measured against honest benchmarks.

Engineering

Python and TypeScript end to end: PyTorch, LightGBM, FastAPI, React and Next.js. Reproducible pipelines, offline test suites, Docker and one-command reruns.

SurgiVision

A surgical-vision research lab for gallbladder surgery. Built with Viraj Rungta. It segments laparoscopic video pixel by pixel, tracks anatomy and instruments from frame to frame, runs an open-source pre-op risk model, and has a simulated smart-glasses voice that reads out what the model sees. Every output is labelled real model, demo or not evaluated. A research prototype, not for clinical use.

Python · FastAPI · OpenCV · React · private repo

Screenshots of SurgiVision: live segmentation of a synthetic laparoscopic clip, and the research workspace
Research figure from the repo: cross-cohort ROC-AUC, reliability and biomarker stability

Does a voice-based Parkinson's classifier survive new patients? Trained on one public cohort, then tested on completely independent ones: 0.88 AUC on one, 0.67 on another, lifted to 0.75 with label-free adaptation. It measures calibration and domain shift, explains predictions with SHAP, and reproduces two popular repos to show where their accuracy comes from data leakage.

Python · LightGBM · SHAP · conformal prediction · source

A study workspace that watches out for you. Courses, assignments, notes, flashcards and a Socratic AI tutor in one place. While you work it reads posture, gaze and blink rate from the webcam and rhythm from your typing, turns them into a wellness score and a burnout forecast, and nudges you to take a break. Camera frames never leave the browser.

Next.js · TypeScript · MediaPipe · Tailwind · source

Screenshots of the Anchor app: wellness overview and burnout forecast
Screenshots of the stock research app: portfolio overview and backtest results

A stock predictor that reports what the models actually did. Built for the Wharton Global Youth investment competition. It scores 74 large-cap US stocks over 1 to 12 months, builds a portfolio for an investor profile, and backtests it over 20 years after costs. The portfolio beat the S&P 500 but lost to simply holding everything equally, and the app says so on the page.

TypeScript · React · Python · FastAPI · source

Water-quality forecasts for US rivers. Started as a physics-informed neural network predicting nitrate risk in Alameda Creek, California, from live stream gauges, weather and satellite imagery. It now forecasts across 359 USGS stations, is validated on stations and states it never trained on, and tracks its own accuracy in a public dashboard.

Python · PyTorch · PINN · USGS data · source

Screenshots of Hydrosense: national station map and the live Alameda Creek transport model
Gameplay recording of Neon Escape: the player collects energy cores while an enemy chases

Grab the cores, dodge the enemy, reach the portal. A maze-escape arcade game written in Python with pygame. Three levels, bombs, teleporters, a dash ability, and enemies that hunt you with A* pathfinding. The clip is a recording of the real game.

Python · pygame · A* pathfinding · source

GitHub contribution streak

Also: pixel art in Python · StangHacks 2026 · all repositories

Popular repositories Loading

  1. Stock-Prediction-Model Stock-Prediction-Model Public

    Wharton Youth Competition

    TypeScript 2

  2. pixel_art pixel_art Public

    This is a pixel art designed on Mario and Luigi:

    Python 1

  3. Neon-Escape Neon-Escape Public

    Escape Room Type Game where player is trying to escape from the enemy catching coins to pass through levels

    Python 1

  4. Hydrosense Hydrosense Public

    Hydrosense is a physics-informed water-quality forecasting project for Alameda Creek, California. It combines live stream and weather data, satellite imagery, and a PINN-based research model to pre…

    Python 1

  5. Anchor Anchor Public

    A smart learning web app for kids that uses AI and computer vision to support studying, provide help, and create a more personalized monitored environment for better learning outcomes.

    TypeScript 1

  6. Parkinson-Model-Detection Parkinson-Model-Detection Public

    Python