Based in Hong Kong · Open to Opportunities

Yash Kat

Developer
Scroll
About Me

Building things
that matter.

I'm a Computer Science student at HKUST with a passion for building products that sit at the intersection of intelligent systems and real-world impact. From ASL recognition to AI navigation, I care about how technology serves people.

Currently interning at BNP Paribas in Global Markets, working on automated market making systems. I've secured a $100K grant, co-founded a startup, and shipped apps used in 3 countries.

AI / ML Full-Stack Systems Engineering Assistive Tech Open Source Quant Finance
3+
Countries Reached
$100K
Grant Secured
2027
HKUST Grad
Projects

Things I've built.

Canasta

Real-Time Multiplayer Card Game

A 2-vs-2 Rajasthani Canasta game with online rooms, solo play against bots, and a dedicated rules engine for melds, wild cards, team reserves, and round scoring.

React TypeScript FastAPI Socket.IO Zustand
GitHub

Agent Runtime Lab

Agent Runtime Research · Foundations

Provider-neutral runtime foundations: typed contracts, validated run-state transitions, scoped default-deny policies, and a local policy-validation API, backed by property and adversarial tests.

Python FastAPI Pydantic PyYAML Hypothesis

PatchWatch + EvalLab

Python Dependency Upgrade Evaluator

An evidence-first dependency upgrade workflow: inspect a Python repo, approve a migration plan, verify repairs in Docker, and review the patch with check logs and independent evaluation.

Python Docker Pydantic SQLAlchemy uv
GitHub

MERIDIAN

Daily Intelligence Terminal

A daily dashboard combining live markets, curated news, weather, job listings, and personal planning. Optional Gemini briefings synthesize the day's signals, with a rule-based fallback.

React JavaScript Express Tailwind CSS Gemini

NexusIDE

Local AI Coding Workspace

A shared coding-agent engine across a VS Code extension, browser dashboard, and CLI. Route tasks across model providers, review proposed edits, approve tests, and inspect local execution traces.

TypeScript React Node.js VS Code Playwright
GitHub

ML Trade Engine

Quantitative Research & ML Backtesting

A Python research pipeline combining XGBoost, LightGBM, and LSTM signals with walk-forward validation, out-of-sample backtests, Monte Carlo simulations, and layered risk controls.

Python XGBoost LightGBM PyTorch MLflow
GitHub
ASL TRIGRAM LM Confidence: 71% ~30 fps ●

ContextSign

Context-Aware ASL Fingerspelling Recognition

A 3-layer real-time system: KNN/SVM classifier over 89-dim MediaPipe hand landmarks, disambiguated by a trigram language model, confirmed via TTS + gesture recognition.

Python MediaPipe scikit-learn NLTK OpenCV
GitHub
Home Living Room Office Storage Shelf A Cabinet Box 1 Box 2 QR Realtime Hierarchical Location Trees · QR Scanning · Knowledge Graph

StorageSync

Cross-Platform Hierarchical Inventory Manager

Full-stack home inventory app with recursive location trees, QR/barcode batch scanning, real-time Supabase sync, force-directed knowledge graph, and Stripe subscription tiers — published on iOS & Android.

TypeScript React Supabase Capacitor Tailwind
GitHub
TRADE SCORE 84.2 Demographics Competition Traffic AI ORACLE 6 Live Data Sources · Retail Site Intelligence

Atlas

AI-Powered Retail Site Intelligence

Pick a location on a map and Atlas ingests live data from 6 external sources — demographics, competitors, parcels, traffic, schools — scores candidate sites, and runs an LLM bull-vs-bear debate to synthesize a recommendation.

Python FastAPI React Supabase OpenAI
GitHub
MORGAN STANLEY ZUBIN FOUNDATION AI CHATBOT Community · Accessibility · Impact HACKATHON

Zubin Foundation Website

Morgan Stanley Hackathon · Community Impact Platform

A Next.js platform built with my team for the Morgan Stanley x Zubin Foundation hackathon — serving Hong Kong's ethnic minority community with accessible resources, translated support, and an AI chatbot layer to connect users with services.

JavaScript Next.js Tailwind Python AI
GitHub
Actual Traffic LSTM Prediction LSTM · 64u MAE 2.84 RMSE 4.12 R² 0.91 time →

Traffic Prediction

LSTM Time-Series Forecasting · Keras + Theano

An experimental deep-learning project using LSTM neural networks to forecast urban traffic flow from time-series sensor data. Built on Keras with Theano backend and MongoDB-stored hyperparameter sweeps for grid optimization.

Python Keras Theano LSTM MongoDB
GitHub
Y R A K CASTLE HP 70% HP · 4 Members · Mending > Ending ⚠ DOOMSCROLLING MINNEHACK 2026

Brainrot Defender

Gamified Group Screen-Time App · MinneHack 2026

A hackathon app that turns accountability into a game: your squad shares a castle whose HP degrades as members doomscroll. Set per-app time limits together and defend the castle — or watch it fall.

React Python FastAPI CSS
GitHub
CCC 8471 BTKI 8471 HS anchor 6-digit Match confidence Top 5 GPT-4o China CCC 2024 · Indonesia BTKI 2022 · WCO HS 2022

TariffMapper

Tsunami Advisors · China ↔ Indonesia AI Tariff Mapping

Next.js prototype that maps customs classification codes between China (CCC) and Indonesia (BTKI/AHTN) using GPT-4o. Search by description, HS code, or national tariff line — ranked matches with confidence, match basis, and source references.

Next.js React OpenAI GPT-4o Supabase
GitHub
Career Goals

Where I'm headed.

I want to build systems that are both technically rigorous and genuinely useful to real people — whether that's quantitative models that move markets, or assistive tools that open doors for underserved communities.

I'm drawn to roles at the intersection of software engineering and intelligent systems — quant development, AI research engineering, and product roles at companies building things that last.

Quant Development AI Research Engineering Systems at Scale Assistive Technology
Documents

Resume & Cover Letter.

Download my latest documents below.