Skip to content
View wheevu's full-sized avatar
0 followers · 1 following
  • Can Tho, Vietnam
  • 20:33 (UTC +07:00)
  • LinkedIn in/wheevu

Block or report wheevu

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
wheevu/README.md

Huy Vu (Josh)

CS Student · AI Engineering · Speech & Language Systems
Can Tho University · Vietnam

I build end-to-end AI systems where language, data, and infrastructure meet.
Speech recognition, NLP pipelines, and ML systems that survive contact with reality (or at least my tests).

🧭 What I Care About

  • 🔊 Automatic Speech Recognition for Vietnamese
  • 🧠 NLP systems with linguistic grounding
  • 🏗️ Data-centric ML & MLOps
  • 🤖 Efficient AI
  • 🌏 Cross-cultural deployment (VN · KR · global)

Former English teacher (IELTS Writing).
I used to debug humans; now I help machines make better mistakes.

🧰 Tech I Actually Use (and occasionally argue with)

🌱 Background

  • CS Major @ Can Tho University (2021–2026)
  • 3+ years teaching English & IELTS Writing
  • IELTS 8.5 overall (chasing 9.0, calmly)
  • Led cross-cultural technical teams (Vietnam ↔ Korea)
  • Big fan of jjajangmyeon (짜장면) and bitter melon soup (canh khổ qua)🍴

📬 Connect

Code compiles better when Huh Yunjin is singing. I think.

Pinned Loading

  1. nemo-vietnamese-asr nemo-vietnamese-asr Public

    End-to-end Vietnamese ASR pipeline using NVIDIA NeMo. Features production-grade CI/CD (GitHub Actions), hardware-aware optimization (40% latency reduction), and robust linguistic data testing.

    Python

  2. episodic-memory-pipeline episodic-memory-pipeline Public

    Local-first agent architecture separating episodic (events) and semantic (facts) memory, with provenance tracking, defense-in-depth LLM sanitization, and multilingual support via Qwen-2.5 + BGE-M3.

    Python

  3. naver-lens naver-lens Public

    Full-stack e-commerce app that enhances the online shopping experience with an integrated AI assistant. Developed for the NAVER Vietnam AI Hackathon.

    TypeScript

  4. repo-to-prompt repo-to-prompt Public

    Tool to export repositories into LLM-friendly context packs and JSONL chunks for RAG (with citations, .gitignore support, redaction, and deterministic output).

    Python