ML / LLM
Models, embeddings, eval
Applied model work for embeddings, ranking, semantic retrieval, and benchmark-driven iteration in production-minded support systems.
PyTorch
Transformers
RAGRerankingEmbeddingsSemantic SearchLLM Evaluation
Software Engineer
LLM Systems, RAG & AI Infrastructure
C++ / Python
I build
AI systems
that earn
trust
in production
Latency, grounding, and reliability decide whether an AI product gets used or ignored. I care about the full stack behind that outcome, from ingestion and retrieval quality to evaluation, model behavior, and the product experience around every answer.
Core focus
LLM Systems
RAG Pipelines
AI Infrastructure
Backend Engineering
Retrieval & Evaluation
Applied ML
From ingestion
to reliable answers
Experience
AdvancedGRC
Software Engineer (AI/ML Systems)
New York City | May 2025 - Present
Primary engineer on a PDF-backed RAG support product, owning ingestion, retrieval, reranking, evaluation, and grounded UX across a FastAPI backend and React/Vite surfaces. Shipped Osmos for daily internal use and cut repetitive support tickets by 36%.
→
DigiPen Institute of Technology Singapore
Teaching Assistant
Singapore | Aug 2023 - Dec 2024
Led labs, debugging sessions, office hours, and grading across programming, software engineering, and math-heavy computer science modules. Helped students build stronger fundamentals through structured 1:1 support and clear evaluation standards.
→
DT Asia Pte Ltd
Software Engineer
Singapore | Jul 2019 - Jul 2020
Converted from intern to full-time engineer, building enterprise log-management and security workflows with Syslog-ng, Azure, and CloudWatch integrations. Also represented the team through live product demos and customer-facing technical presentations.
→
Projects
Tech stack
Inference / Retrieval
Serving, search, confidence
Serving and retrieval tooling for grounded answers, multi-KB search, vector indices, and confidence-aware answer routing.
Ollama
FAISS
CUDA
Confidence Gating
Backend / Deployment
APIs, jobs, persistence
FastAPI services, ingestion pipelines, containerized local development, and durable data flows for AI products that need to ship cleanly.
FastAPI
Docker
SQLite
Linux
Git
Languages / Frontend
Core languages, UI delivery
Core implementation languages plus the frontend tools I use to ship full-stack AI interfaces, widgets, and internal engineering tools.
Python
C++
C
C#
React
Vite
Contact
Let's
talk.
Open to new-grad AI/ML and software engineering roles from June 2026. If you're hiring for LLM systems, backend infrastructure, or applied ML work, I'd be glad to talk.