- You are subscribed to this talk through .
- You are watching this talk through .
- You are subscribed to this talk. (unsubscribe, watch)
- You are watching this talk. (unwatch, subscribe)
- You are not subscribed to this talk. (watch, subscribe)
Sorry! I did not realize that this talk was changed to CSI 2127 (the Computer Science Instructional Center, next to the Iribe Building.)
This workshop combines the best of two engaging sessions, offering a comprehensive and hands-on experience in building advanced AI applications and AI agents. We will use the state-of-the-art Large Language Models (LLMs) and Vision Language Models(VLMs) to explain the process. Participants will explore practical techniques to leverage these powerful models for tasks such as Natural Language Understanding (NLU), Natural Language Processing (NLP), decision-making, text/image generation, and task automation. This workshop enables attendees to understand the essential topics, including open-source package requirements in Python, model selection, fine-tuning, and integrating external data sources. The session covers the design and implementation of AI agents by utilizing open-source models like Llama for tuning parameters, and deploying real-world AI solutions. Emphasis will be placed on architecture design, training optimization, scalability, efficiency, and ethical considerations. The workshop will adapt its use case and model selection (e.g., language agent vs. vision agent) based on the audience's background to ensure relevance and effectiveness of the session. By the end of the session, participants will have the foundational knowledge and skills to effectively create and deploy full-stack AI applications and agents tailored to their fields of interest.
Pizza is available for those who register!
Note that this is an ACM talk (the ACM Distinguished Speakers program), aimed at advanced undergraduates, but anyone is welcome to attend. In particular, graduate students without background in the subject may be interested.
Dr. Mehdi Bahrami is a Senior Member of both ACM and IEEE. He is a Principal Researcher at Fujitsu Research of America in California. With expertise in Generative AI, Applied Machine Learning, and diverse API integration at scale, his work focuses on advancing cutting-edge AI technologies. He holds a Ph.D. in Electrical Engineering and Computer Science from the University of California, Merced. Dr. Bahrami has over 15 years of software industry experience, complemented by more than five years of academic engagement, all toward contributing to AutoML, Natural Language Processing, and Generative AI.
Dr. Bahrami is a recipient of several awards, such as the 2024 IEEE Outstanding Engineer Award for his “pioneering contributions to generative AI and API automation”, the 2024 Fujitsu Research Group Head's Award for “achievements in AI trust technologies”, and the 2016 ACM ICN Best Demo Award. He has also received prestigious fellowships and leadership awards during his doctoral studies, such as the Fletcher Jones Fellowship and the Distinguished Leadership Award. Dr. Bahrami is the author of over 30 publications and the inventor of more than 34 granted U.S. patents. His work has been featured in prominent media outlets, including MIT Technology Review.
Dr. Bahrami has contributed to prestigious academic events through roles such as a chair, editor, and reviewer for several international conferences and journals including ACL, AutoML and COLING. He has served as an AI panelist for the National Science Foundation’s Small Business Innovation Research (NSF SBIR) program. Dr. Bahrami has delivered numerous invited talks and tutorials at prestigious international conferences, universities, and industry events. His presentations span diverse topics such as API Integration, Generative AI, and AutoML. Notable engagements include keynote tutorials at ACM/IEEE conferences, lectures at Stanford / Carnegie Mellon Universities, and keynote talks at industry summits like the NLP Summit and MLConf. These talks highlight his role in reducing the gap b