My name is Tianyi Huang (Tony), and I’m an AI researcher focused on building trustworthy, reliable NLP systems. Specifically, I work on agentic systems, data‑processing pipelines, retrieval‑augmented generation (RAG), and large language models (LLMs). In parallel, I conduct HCI research to create accessible AI tools for policymakers and educators, advancing AI literacy and providing open‑source resources that empower students worldwide. Whether it’s simplifying everyday tasks or opening up entirely new possibilities, I’m dedicated to using technology as a force for positive change.

• ML Researcher, Cleanlab (2025 – Present)
  Researching data‑centric methods that increase the reliability of LLM pipelines.
• Student Ambassador & Curriculum Developer, App Inventor Foundation (2024 – Present)
  First Student Ambassador; design open‑source curricula for AI literacy and AI tools for educators and policymakers.
• Founder & President, App‑In Club (2022 – Present)
  Lead a global nonprofit with 20 + chapters and 10,000 + learners, offering free AI/app workshops, events, and resources.
• LLM Research Engineer, Mathos AI (2024 – 2025)
  Built multi‑agent pipelines with advanced calculus solvers, automated data retrieval workflows, and led the TI-84 emulator feature.

• Factual Context Validation & Simplification – Exploring RAG pipelines that compress storage while maintaining/boosting factual accuracy on QA tasks.
• Structured Reasoning for Fairness – Multi‑agent workflow that disentangles fact vs opinion and flags textual bias with interpretable scores.
• Cost‑Effective Robotic Handwriting System – $56 Raspberry Pi plotter reproducing user handwriting at ±0.3 mm precision.
• Synthetic Data for HAB Detection – Investigates Gaussian‑copula augmentation and explores optimal synthetic‑data ratios for harmful algal bloom detection.
• More Details → Google Scholar

Feel free to reach out on GitHub, LinkedIn, or Email via the buttons below!