My Headshot

Hi, I’m Jingtao Scott Hong

Graduate student in Business Analytics at Columbia University with a Computer Science background from the University of Virginia. Skilled in AI, data science, and software engineering, with a focus on developing robust, scalable solutions. Experience includes building AI-powered automation platforms, optimizing data architectures, and creating full-stack web applications in roles at HireBeat and KPMG Shanghai. Led projects such as an Azure OpenAI-driven environmental compliance assistant, utilizing advanced NLP, graph databases, and embedding techniques. Proven ability to combine machine learning, data engineering, and software development to create impactful, data-driven business solutions.

Experience

HireBeat

Data Scientist and Artificial Intelligence Engineer Intern
New York, NY | Jun 2024 – Present

  • Spearheaded the development of a Python-based LLM workflow automation platform, featuring integrations with leading APIs such as Google and Microsoft.
  • Architected sophisticated data structures, refined LLM prompts, and implemented benchmarking and testing protocols, building databases with SQL and PostgreSQL.
  • Engineered an advanced Graph Database utilizing a RAG structure on AWS, significantly enhancing data retrieval efficiency and accuracy.

KPMG Shanghai

M&A Advisory Associate Research Intern
Shanghai, China | Jun 2023 – Aug 2023

  • Conducted industry research across sectors like pharmaceuticals and 3D printing, identifying over 100 companies for analysis.
  • Authored white papers detailing market trends and financial performance.
  • Participated in client interviews, contributed to crafting questions, and summarized discussions for M&A advisory support.
  • Prepared presentation decks summarizing research and interview insights for client engagements.

UVA DevHub

Machine Learning Engineer and Software Development Intern
Charlottesville, VA | Jun 2022 – May 2023

  • Developed a full-stack web application for clients using Django, ReactJS, and Bootstrap, incorporating features like keyword research.
  • Enhanced performance and security for the client’s website through interface improvements and online data storage integration.
  • Utilized TensorFlow to create predictive pricing and maintenance models, reducing IT downtime by 9%.
  • Engineered and deployed an automated data pipeline using AWS Lambda and Python, optimizing data processing times and enhancing data management efficiency.

Research & Projects

EcoLegal Assistant: LLM-Driven Environmental Compliance Solutions

New York, NY | Jan 2024 – Apr 2024

  • Capstone Project Leader with GHD INC.
  • Pre-processed a comprehensive dataset derived from Title 40 of the Code of Federal Regulations and web-scraped 1M+ text, graphs, 100+ tables, and 2000+ formulae as relevant information, storing in a Neo4j graph database.
  • Designed and deployed an Azure OpenAI-powered LLM solution for environmental compliance, using sophisticated prompt engineering, embedding matching, and Cypher querying.
  • Conducted 100+ experiments to fine-tune 10+ hyperparameters, implementing advanced techniques in NLP, embeddings, and graph-based querying.

Machine Learning Research: SoK: Pitfalls in Evaluating Black-Box Attacks

Charlottesville, VA | Aug 2022 – Apr 2024

  • Group Research advised by Professor David Evans & Yuan Tian of UVA.
  • Designed a comprehensive platform that reproduces 30+ SOTA adversarial attacks, implementing 10+ kinds of adversarial attacks (3000+ lines of code).
  • Paper accepted by the 2nd IEEE Conference on Secure and Trustworthy Machine Learning.
  • Link to Paper

Leadership Experience

Mainland Student Network | External Vice President

Charlottesville, VA | Oct 2019 – May 2023

  • Outreached and maintained connections with over 20+ organizations for events and organizational funding.
  • Led the organization’s publicity efforts, managing 3+ social media accounts and providing high-level guidance.
  • Supported logistics, outreach, and publicity for 20+ events, developing and maintaining 8 sponsorship relations.