Experience

Work and research

1. Oct 2025 – Present, Atlanta, GA

   Undergraduate Researcher, with Prof. Taesoo Kim

   Systems Software & Security Lab, Georgia Tech

   • Engineered an LLM-driven security testing framework using shallow bugs as coverage signals to explore deeper execution paths, with automated patching and evaluation across databases, networking tools and bootloaders.
   • Identified 11 security vulnerabilities, including 5 rated CVSS 8.0+, in open-source software assessments.
   • Integrated ATLANTIS, an AI-driven vulnerability discovery system, into Docker-based CI/CD, using call graphs to trace callers and dependencies of changed functions, generate targeted fuzzing harnesses and compare coverage of new or modified code against existing harnesses.

   Research →

2. Jul 2026 – Sep 2026, Kuala Lumpur, Malaysia

   Lead Software Engineer

   AI.SEA (Eupe Corporation Berhad)

   • Engineered an end-to-end agentic Bill of Quantities (BQ) system, transforming messy, parser-unfriendly Excel workbooks into hierarchical, classified construction-cost data with 100% financial reconciliation across $148M.
   • Developed multi-agent schema discovery with OpenAI Agents SDK, deterministic validation, sandbox analysis and adversarial review, cutting manual schema creation from seven weeks to one overnight run.
   • Built a FastAPI, DuckDB, React and Docker production stack with parallel processing, human review, immutable audits and versioned pricing snapshots to preserve data integrity across concurrent workloads.

   Arkiba AI →

   

3. May 2026 – Jul 2026, Hong Kong

   Data Science Intern

   Queen Mary Hospital, HKU

   • Architected a full-stack medical AI application for an osteoporosis prevention study involving 10,000+ patients, integrating a RAG-powered WhatsApp chatbot for DEXA scan intake, patient data entry and analysis.
   • Developed a PyTorch/OpenCV model to detect femoral malrotation during intramedullary nailing, supporting real-time intraoperative surgical decision-making.
   • Engineered an OpenCV volumetric segmentation pipeline for MRI and CT images and a gesture-controlled analyzer, enabling touchless intraoperative review with less manual image handling.

   

4. Jan 2025 – Aug 2025, Kuala Lumpur, Malaysia

   Co-founder & CTO

   Byte In (gamified AI calorie-tracking startup)

   • Built a computer-vision ML pipeline with A/B testing, achieving 98% food-recognition accuracy (23% relative improvement over baseline) and serving 1,000+ daily inferences.
   • Architected Python, Django and PostgreSQL APIs and a PyTorch-to-ONNX pipeline for training and low-latency model serving; deployed the production server in Docker on DigitalOcean.
   • Secured $23K in competitive grant funding from Cradle Fund under Malaysia’s Ministry of Finance.

5. Jun 2024 – Jan 2025, Kuala Lumpur, Malaysia

   AI Intern

   University of Malaya

   • Deployed a production LLM system designed for 30,000+ users, integrating RAG, a vector database and prompt engineering to automate Q&A, course registration and scheduling.
   • Built a multi-agent architecture with intent routing, tool integrations and guardrails to handle Q&A, course-registration and scheduling requests.
   • Developed Django REST/PostgreSQL APIs and a React/TypeScript frontend supporting course-registration and scheduling workflows, integrating LangChain for LLM orchestration.

6. Jun 2023 – Aug 2023, Remote

   Junior ML Engineer, Model Evaluation Task Lead

   Omdena Algiers Local Chapter

   • Led the Model Evaluation Team, building reproducible benchmarks across 10+ ML/transformer architectures on 13.5K Algerian-dialect texts using accuracy, precision, recall and F1.
   • Proposed and fine-tuned two multilingual models; standardized preprocessing and stratified 80/10/10 train/validation/test splits with data engineering and model development teams.
   • Co-authored the paper for the North Africans in Machine Learning (NAML) workshop, NeurIPS 2023; benchmark results favored DzaraShield (0.87 F1) over multilingual E5 (0.80) and SBERT (0.77). Paper cited in six publications.

   The paper →

   

1. Expected May 2028, Atlanta, GA

   BS Computer Science

   Georgia Institute of Technology

   • Threads: Intelligence and Devices
   • GPA 4.0. Faculty Honors.
   • Coursework: Data Structures & Algorithms, Machine Learning, Systems & Networks, Perception & Robotics, Calculus