Building the future of artificial intelligence at Kolosal AI. Researcher at UBC, where I teach machines to decode biology — from circular RNA to tumor suppressor genes — and occasionally, the stock market.
Open-source platform for running large language models locally — private, fast, and fully under your control. Custom training and production-ready inference at scale.
Published ANN pipeline classifying circRNA-disease associations with Gaussian blur preprocessing — 75% accuracy at 0.14ms per prediction.
Production-ready inference with TensorRT acceleration. Handles batching, model versioning, and GPU memory management out of the box.
End-to-end AutoML with Gradio UI — from hyperparameter tuning with Optuna to experiment tracking with MLflow, all in one platform.
Industrial-grade Self-Organizing Map and clustering library for Rust — published to crates.io with CPU, CUDA, and Metal backends. Supports SOM classification, KMeans variants (KMeans++, KDE-KMeans, SOM++), and automatic model selection via silhouette scoring.
Real-time pose estimation with MediaPipe for form correction, rep counting, and personalized workout recommendations.
High-performance nuclear chain reaction simulator in Rust — GPU-accelerated particle physics with real-time 3D visualization, 150+ isotope library from ENDF/B-VIII.0 nuclear data, and full neutron genealogy tracking.
Desktop app for evaluating stocks through technical and fundamental analysis, powered by on-device ML inference and a local Qwen2.5 LLM — fully offline, no cloud required.
GAT-VGAE + SOM pipeline analysing 249,455 ZINC15 molecules with counterfactual QED decomposition — mapping functional group enrichment across drug-like chemical space in Rust.
Pure-Rust two-tower MLP learning entity embeddings and cell-type expression profiles to predict TF–gene interactions — 83.06% ensemble accuracy, CPU-trainable, no deep learning framework.
Diagnosed three critical gradient failures in two-tower GRN models, then compared corrected modular and monolithic cross-encoders — cross-encoder achieves AUROC 0.904 vs 0.810, with far greater imbalance robustness.
Parameter-matched comparison (5.58M params) of dual-encoder and cross-encoder architectures across four imbalance regimes, ablation, pruning, and cold-start evaluation — submitted to IEEE TNNLS.
Unsupervised clustering of 190 countries over 45 years reveals four distinct economic paths — from stagnation to exponential growth.
Do crypto markets lead or lag equities around recessions? Cross-correlation and Granger causality analysis across five business cycles.
Optimal BTC sizing via Risk-Budget Framework. Component Risk Contribution analysis across five portfolio profiles finds 10–12% is the evidence-based ceiling.
Community-driven platform connecting learners and researchers through structured courses and peer mentorship programs.
Building mental health awareness across Indonesia through accessible psychology content and peer-to-peer support networks.
Undergraduate Researcher
Building deep learning pipelines for circular RNA classification and genomic sequence analysis.
Co-Founder
Architecting open-source tools for local LLM deployment, inference optimization, and MLOps automation.
First Author
"Deep Learning Algorithm with Gaussian Blur Data Pre-processing in Circular RNA Classification"
I chase problems where data hides something meaningful — a motif buried in RNA sequences, a leading signal in financial time series, or a chance to make ML infrastructure less painful.
Currently an undergraduate researcher at UBC building deep learning systems for genomic data, and co-founding Kolosal AI to make running LLMs locally actually simple. My first-author publication on circular RNA classification is live in URNCST Journal.
When I'm not debugging loss curves, I'm reading about macroeconomics or exploring Vancouver's trails.