San Diego, California, United States
Hello! My name is Jason, and I recently earned a B.S. in Data Science from UC San Diego. I enjoy working with data to find practical solutions and improve how things work in both tech and business settings. I have experience with Python, SQL, Tableau, and AWS and I’m looking to apply those skills in a full-time role. Please feel free to connect with me here or contact me at: Email: [email protected] Phone: (626)-277-8952 Website: https://jingchenggu.github.io/ GitHub: https://github.com/JingChengGu
I run both the AI engineering and operations side of a multi-country deployment — building and shipping LLM-powered automation systems in production while managing the 50+ person organization running on top of them. At an early-stage startup, there's no handoff between "who builds it" and "who owns the outcome." - Designed and shipped custom GenAI integrations and AI agent workflows using LangChain, Dify, Python, and REST APIs — wiring LLMs to live call data, CRM systems, and operational pipelines across 50+ agents in 3 countries - Built output evaluation frameworks tracking field-level accuracy, hallucination rates, and adoption signals across concurrent production deployments — iterated on prompt engineering and integration logic based on live results to continuously improve system reliability - Oversaw deployment of ASR and LLM-assisted workflow tools across live call center operations — managed agent adoption, QA feedback loops, and real-time troubleshooting for 10+ SMB clients across mortgage, solar, insurance, and real estate - Stepped into B2B sales after director departure — identified a mortgage client's pain point, scoped the automation solution, closed the deal solo, and doubled their revenue within 3 months - Manage 4 team leads, 1 ops director, and 50+ BPO agents across Philippines, Mexico, and Bolivia — owning both AI system performance and the operational outcomes that depend on it
- Analyzed EV charging patterns to identify underserved regions across San Diego. - Forecasted EV adoption trends to guide infrastructure planning through 2025. - Built geospatial dashboards to support charger placement strategy. - Developed image segmentation + classification pipeline to detect charger faults. - Designed and demoed a mobile app concept to streamline charger issue reporting. - Presented solutions to 200+ employees, including engineers and directors.
'The Data Science Student Society at UCSD is an interdisciplinary academic organization designed to immerse the community in the diverse and growing facets of Data Science: Machine Learning, Computational Statistics, Data Mining, Visualization, Predictive Analytics, and any new emerging relevant fields of study. With practical hands-on data projects, a professional portfolio-building approach, and fun outreach activities, the Data Science Student Society at UCSD strives to enrich the academic life of the student community by strengthening them for success in their current and future pursuits of Data Science related fields.'
- Leveraged Excel's data visualization tools and SQL-driven insights to optimize the alumni outreach strategy, leading to a 25% increase in alumni participation, bolstering chapter events, and strengthening the organization’s network. - Established and maintained a comprehensive database of 300+ alumni, ensuring accurate and easily accessible records. - Utilized SQL queries to segment and analyze alumni data, leading to more targeted communications and improved engagement.
- Built QA dashboard to monitor real-time device performance across 50+ units. - Streamlined ETL pipeline to surface temperature and image capture issues. - Detected anomalies in image confidence scores using AWS QuickSight. - Enhanced model retraining efficiency by 30% through data relabeling and annotation.