AI After Hours @ GitHub w/ Encord
Join us at GitHub HQ - connect with fellow AI enthusiasts, learn from industry experts, & enjoy a slice of pizza 🍕
Agenda
5:30pm: Doors open
6pm - 8:00pm: Lightning talks ⚡
8:00pm - 8:30pm: Networking
Talks
⚡ Talk 1: Leveraging LLM Based Marketing Analytics Copilots
Speaker: Sai Kumar Arava, Machine Learning Manager @ Adobe
Abstract: This talk explores the use of large language models (LLMs) to simplify queries around marketing analytics. It highlights the challenges of using LLMs for domain-specific question-answering, SQL generation, and tabular analysis, and demonstrates how techniques like semantic search, prompt engineering, and fine-tuning can enhance model performance.
⚡ Talk 2: AI-Powered Fundraising for Conservation: Transforming Grant Discovery
Speaker: Prajakta Pardeshi, Sr Machine Learning Scientist @ Walmart Global Tech
Abstract: Discover how AI is revolutionizing the way conservation non-profits identify and secure grants, with automated tools that streamline research and maximize funding success.
⚡ Talk 3: Multimodal RAG for video
Speaker: Anup Gosavi, CEO & Co-founder @ VideoDB
Abstract: Large Language Models (LLMs) can summarize, search & personalize text for each user but fall short in doing the same for video, especially recorded video. A simple query like "show me highlights of last night's football game" goes unfulfilled. That's because while constructing a RAG pipeline for text has now become relatively straightforward, there are challenges of adapting RAG for video content which include not only video intelligence but also streaming the videos in real time. This talk talks about approaches and infrastructure change needed to create multimodal RAG for video
⚡ Talk 4: Personalized Video Ads at Scale with GenAI
Speaker: Shradha Agrawal, Engineering Manager - GenAI, Computer Vision @ Adobe
Abstract: Discover how generative AI is revolutionizing video advertising. In this talk, we'll explore GenVideo, a cutting-edge technology that enables one-shot, target-image and shape-aware video editing using Text-to-Image (T2I) diffusion models. Learn how this innovation allows for rapid creation of personalized video ads at scale, potentially transforming your advertising strategies and workflow.
⚡ Talk 5: Unleashing Potential: From POCs to a Promising Future in Agentic AI
Speaker: Meghana Puvvadi, Director of Engineering AI/ML Enterprise Assistants @ NVIDIA
Abstract: This talk will provide a comprehensive overview of AI agents, exploring fundamental concepts, prompt engineering strategies, and Retrieval-Augmented Generation (RAG). The presentation delves into real-world applications of agentic AI while addressing the challenges of bringing these technologies to production environments.
Hosts
Encord - Securely consolidate data from multiple sources into a single platform to efficiently curate and annotate high-quality datasets for developing production-grade AI at scale