Thanh Dat Le
***.************@*****.*** +84-329-***-*** (VN) +1-857-***-**** (US) github.com/lethanhdatphumy Education
Northeastern University, USA Boston, MA
B.S. & M.S. in Computer Science (Combined Program) May 2027
• GPA: 3.85 Coursework: Machine Learning, Deep Learning, Algorithms, Object Oriented Programming, Database Management. Technical Skills
Languages: Python, Java, SQL, C/C++, R
AI/ML: LLM Fine-Tuning (LoRA/Unsloth), Gemma 2, PyTorch, LangChain, RAG, FAISS, Computer Vision Backend/Cloud: Spring Boot 3, FastAPI, Docker, AWS (ECR, S3), GitHub Actions (CI/CD), DVC Data/NLP: PostgreSQL (pgvector), MySQL, Vector Embeddings, BERT, PyMuPDF Research Experience
AIMA Research Mentorship Program Remote
AI Research Mentee Jan 2026 – Present
• Selected for a competitive AI research mentorship program focusing on applied AI methodologies.
• Conducting research under the mentorship of a PhD candidate at Carnegie Mellon University (CMU).
• Investigating advanced computer vision techniques using Meta’s Segment Anything Model (SAM). Projects
Animals Detection MLOps System PyTorch, AWS ECR, DVC GitHub 2026
• Built a production-grade Computer Vision pipeline for image classification using PyTorch and Transfer Learning.
• Implemented an end-to-end MLOps workflow with DVC for orchestration and GitHub Actions for CI/CD, automatically pushing Docker images to AWS ECR.
• Deployed a robust FastAPI inference server with Out-of-Distribution (OOD) handling to flag low-confidence predictions. Clinical Diagnosis RAG System (Fine-Tuned LLM) Gemma 2, LoRA, Docker, AWS GitHub 2025
• Fine-Tuned the Gemma 2 9B model using LoRA and Unsloth, significantly enhancing clinical reasoning capabilities compared to base models.
• Engineered a RAG pipeline achieving 90% diagnosis accuracy and 73.1% keyword match by combining the fine-tuned LLM with FAISS semantic search.
• Containerized the application using Docker for consistent deployment and managed data versioning via DVC/AWS S3.
• Deployed a user-facing interactive diagnostic interface using Streamlit. Cocktail Rec. Engine FastAPI, pgvector GitHub 2025
• Built a semantic search engine using SentenceTransformers and PostgreSQL pgvector for NLP queries.
• Deployed containerized FastAPI backend via Docker for real-time cosine similarity search.
• Integrated LLM re-ranking to significantly boost recommendation relevance and user engagement. Campus Food Delivery Spring Boot 3, AWS S3 GitHub 2024
• Architected Spring Boot 3 backend with Liquibase versioning and Hibernate Envers auditing.
• Secured API with Spring Security, JWT, and brute-force protection; integrated AWS S3 for storage. Leadership & Academic Experience
Graduate OOD Game Project Northeastern University
Team Lead 2025
• Led a 5-member agile team, enforcing SOLID principles and UML to ensure code maintainability.
• Managed GitHub CI/CD workflows, delivering a stable game prototype with zero critical bugs. DSA Study Group Ho Chi Minh City
Founder & Team Lead Sep 2024 – Jan 2025
• Founded a peer learning cohort for technical interview prep, guided by a Senior Microsoft Engineer.
• Taught Algorithmic Complexity (O(n) vs O(n2)), improving group problem-solving speed by 40%. Certifications
Mathematics for ML: Calculus & Linear Algebra – DeepLearning.AI