My current main research interests lie in AI for Software Development (AI4Code, AI4Software), as well as Large Language Models and Agentic Systems. I'm currently conducting research under the guidance of Prof. Bach Le and Dr. Nghi D. Q. Bui at FPT Software AI Center. I graduated early with an excellent Bachelor's degree in Computer Science from Hanoi University of Science and Technology. My research focuses on developing algorithms to train and fine-tune Large Language Models for code, and exploring agentic systems for Software Engineering tasks.

Positions: I am currently looking for PhD opportunities in Computer Science for the next academic cycle, with research interests in AI for Software Engineering, particularly foundation models for code, autonomous coding agents, and long-horizon software evolution.

Funding: I have been awarded the Techcombank Overseas PhD Study Sponsorship, providing up to USD 333,000 for overseas PhD study. [Offer Letter (PDF)]


News

  • Nov 2026: Our paper "GeoDPO: Geometric Token Importance for Stable LLM Alignment" has been accepted at EACL 2027!
  • Aug 2026: Our paper "Beyond Repository Boundaries: Cross-Repository Graph Retrieval for Code Generation" has been accepted at Findings of EMNLP 2026!
  • May 2026: Released RustPrint, a documentation-guided agentic framework for migrating large-scale C codebases to Rust while preserving architectural intent.
  • Apr 2026: Our paper "CodeWiki" has been accepted at Findings of ACL 2026!
  • Jan 2026: Converged has been awarded First Prize at AI for Real Action 2025 - Vietnam's first nationally televised AI competition! This year's challenge focused on building a Vietnamese-native language model that balances capability (Tài) and ethics (Đức) for real-world application in high school education. Our product, ConnectED (Connect Education), enables teachers to generate visual simulations and educational videos in minutes from simple descriptions. Special thanks to teammates Tịnh Lương Sơn and Thang Doan!
  • Oct 2025: Released CodeWiki, an open-source framework for AI-powered repository-level documentation across 7 programming languages.
  • Jan 2025: Our paper "Mutual-pairing Data Augmentation for Fewshot Continual Relation Extraction" has been accepted at NAACL 2025!
  • Sept 2024: Our paper "Preserving Generalization of Language Models in Few-shot Continual Relation Extraction" has been accepted at EMNLP 2024!

Publications

  • Secure Agentic Coding through Counterexample-Grounded Feedback (FALCON)
    Tue Le, Anh Nguyen Hoang, Gabriel Sasseville, Doseok Jang, Pierre-Luc Bacon, Nghi D. Q. Bui.
    Under review
  • Can Agentic Code Review Catch Vulnerabilities Before They Are Merged?
    Yihan Yang, Chenyuan Yang, Zijie Zhao, Anh Nguyen Hoang, Ziqi Zhang, Weidong Wang, Zhiheng Lyu, Lingming Zhang.
    Under review
  • Beyond Repository Boundaries: Cross-Repository Graph Retrieval for Code Generation
    Minh Le-Anh, Nam Le Hai, Quyen Tran, Anh Nguyen Hoang, Linh Ngo Van, Bach Le, Nghi D. Q. Bui.
    Findings of EMNLP 2026
  • GeoDPO: Geometric Token Importance for Stable LLM Alignment
    Dat Phi Van, Anh Nguyen Hoang, Linh Ngo Van, Thien Huu Nguyen, Thanh-Toan Do, Thanh Hong Nguyen, Trung Le.
    EACL 2027
  • Enhancing Few-shot Continual Relation Extraction via Pairwise Augmentation and Guided Sharpness-Aware Optimization
    Hoang Thanh Dat*, Nguyen Hoang Anh*, Nam Le Hai*, Linh Ngo Van, Sang Dinh. (*Equal contribution)
    Artificial Intelligence (AIJ), Vol. 362, 2027
  • Documentation-Guided Agentic Codebase Migration from C to Rust Stars
    Minh Le-Anh, Anh Nguyen Hoang, Bach Le, Nghi D. Q. Bui.
    Preprint
  • CodeWiki: Evaluating AI's Ability to Generate Holistic Documentation for Large-Scale Codebases Stars
    Anh Nguyen Hoang, Minh Le-Anh, Bach Le, Nghi D. Q. Bui.
    Findings of ACL 2026
  • Mutual-pairing Data Augmentation for Fewshot Continual Relation Extraction Stars
    Nguyen Hoang Anh*, Quyen Tran*, Thanh Xuan Nguyen*, Nguyen Thi Ngoc Diep, Linh Ngo Van, Thien Huu Nguyen, Trung Le. (*Equal contribution)
    NAACL 2025
  • Preserving Generalization of Language Models in Few-shot Continual Relation Extraction Stars
    Quyen Tran*, Nguyen Xuan Thanh*, Nguyen Hoang Anh*, Nam Le Hai, Trung Le, Linh Van Ngo, Thien Huu Nguyen. (*Equal contribution)
    EMNLP 2024

Open Source Projects

  • XTTSv2-Finetuning-for-New-Languages Stars
    Repository for fine-tuning XTTSv2 (speech synthesizer model) on new languages by extending vocabulary and modifying code in the original Coqui AI TTS project.
  • vnhtr Stars
    Pioneered one of the first attempts to tackle Vietnamese Handwritten Text Recognition, building upon TrOCR with an integrated Rethinking Head on top of the decoder module.

Funding

  • Techcombank Overseas PhD Study Sponsorship
    Recipient, April 2026. Sponsorship of up to USD 333,000 for overseas PhD study. [Offer Letter (PDF)]

Contact