Tien Pham

Cognitive Science & Computer Science at UC Berkeley.
Proficient in software development, AI systems, and technical project management.

I am a senior at UC Berkeley majoring in Cognitive Science with a minor in Computer Science (Expected December 2026), pursuing interdisciplinary work across AI-adjacent software development and human-computer interaction research.

Languages: Python, TS/JS, Java, C#, C++, x86 Assembly (MASM), HTML/CSS, GML
Tools & Platforms: VS Code, Git, GitHub Actions, Docker, Vercel, Hugging Face, Bash, Nginx, Unity, Blender

Projects

Puffin: Browser-Based Drone Simulation Interface | Website, Paper | Figma, UX Research, Usability Testing • Co-designed a browser-based interface for a drone simulation environment, translating complex ROS2/PX4 (simulation) and Docker workflows into intuitive UI workflows using Nielsen's heuristics
• Facilitated 4 need-finding interviews with drone club leaders and hobbyist FPV pilots, and ran a 5-participant think-aloud usability study to evaluate system readiness feedback, control prerequisites, and takeoff completion metrics
• Co-authored the project's 9-page ACM CHI conference-style paper, slide decks, and documentation detailing our team's design process
 Hybrid RAG Search Pipeline & LLM Judge | Python, LangChain, ChromaDB, Groq, Gradio • Built an interactive hybrid RAG prototype in Gradio, combining BM25 keyword search with vector embeddings via a LangChain EnsembleRetriever
• Extended EnsembleRetriever with query rewording and an LLM-as-a-judge prompt to evaluate document relevance, filtering retrieved context prior to generation with Llama 3 8B (via Groq API)
• Configured dynamic UI controls for retrieval weight adjustments (k-search parameters and prompt toggles) to rapidly test Q&A response quality
 SQLite Socket Client/Server | Python, SQLite3, Sockets, Multithreading, Multiprocessing • Built a custom client and concurrent server implementation using Python's native socket module and ThreadPoolExecutor to handle client requests, chunked payload streaming, and parsing of raw UTF-8 JSON byte streams
• Integrated a modular SQLite backend via a dedicated DatabaseManager using WAL mode, thread locks, and a thread-safe connection pool queue to handle concurrent SQL queries without database contention
• Observed that while GIL acts as a bottleneck preventing parallel CPU execution in multithreaded Python, multithreading offers similar performance to multiprocessing with lower memory overhead for I/O-bound socket and database tasks
 The Facility: 3D Indie Action Game | Itch.io Demo | C#, Unity, Blender, Agile/Scrum • Solo-developed and published a 3D first-person game demo on Itch.io as a capstone project, operating under industry mentorship
• Managed end-to-end production using Agile/Scrum frameworks, delivering regular milestone presentations, and maintaining appropriate documentation
• Programmed player movement physics, combat loops, and level mechanics in C# with Unity while modeling 3D assets in Blender

Experience

2025–Present | MiraCosta College | STEM Tutor • Tutor for Calculus I, Biology I, Chemistry I, Python, Java, C++, Data Structures, and Assembly; nominated by faculty for exceptional academic performance
• Designed individualized study strategies to break down complex algorithmic and mathematical concepts and improve long-term student outcomes
 2025–2026 | De Anza College | CIS Teaching Assistant • TA for Python I, Python II, C++ I, and Data Structures; nominated by faculty for exceptional academic performance
• Managed course discussion forums and advised students on how to debug their work and conceptually approach complex algorithmic problems

Education

Expected December 2026 | University of California, Berkeley • Major in Cognitive Science, Minor in Computer Science
• Relevant Coursework: CS 61 series (Structure & Interpretation of Computer Programs, Data Structures & Algorithms, Machine Structures), COGSCI 131 (Computational Models of Cognition), CS 188 (Intro to AI), CS 160 (User Interface Design and Development), and CS 168 (Intro to the Internet)
 2024–2025 | De Anza College • Relevant Coursework: Python III (Advanced), C++ II (Intermediate), Data Structures, and Intro to x86 Assembly
 Certifications • Fundamentals of MCP & AI Agents (Hugging Face)
• Prompt Engineering (Vanderbilt)
• Intro to GenAI (Google Cloud)
• Intro to AI & ML with Python (IBM)
• Intro to TensorFlow (DeepLearning.AI)
• XO Platform (Kore.AI)
• Certificate of Achievement: Programming in Python (De Anza College)