Sunil Pai from Cloudflare discusses building efficient AI agent architectures using Durable Objects and Dynamic Workers as an alternative to platforms like Anthropic's. He explores the search for a standardized 'React-like' framework for agents, the culture of forking in open source, and encourages developers to pursue original, 'sci-fi' style projects.
Merve Noyan from Hugging Face discusses how open-source models have achieved parity with closed-source counterparts, highlighting the Hugging Face ecosystem built to support this shift. She covers tools for model selection, local agent deployment, and the transformative "Hugging Face Skills" that allow agents to automate complex ML engineering tasks like fine-tuning models with a single prompt.
Filip Makraduli from Superlinked discusses the common infrastructure gaps and profiling mistakes encountered when deploying small embedding and transformer models. He introduces the Superlinked Inference Engine (SIE), an open-source solution designed for dynamic model loading, hot-swapping, and memory-aware eviction to maximize GPU utilization and streamline the path from development to production.
Dotta, the creator of Paperclip, introduces it as an open-source orchestrator for building "zero-human companies." This talk demonstrates how to set up an organization of AI agents, leverage skills and custom instructions for reliable work, and automate business processes. Through a live demo, Dotta showcases creating a company from scratch, managing agent workflows with QA and routines, and outlines the exciting future roadmap for the platform.
Peter Steinberger, creator of OpenClaw, discusses his journey into open source, the unexpected success of his AI agent, and how his development workflow has been fundamentally transformed by using agentic AI tools like Codex.
Delivery Hero presents "Asya", an open-source framework that replaces traditional AI pipelines with a distributed, asynchronous actor model. This paradigm shift dramatically lowers GPU costs and improves scalability by treating each processing step as an independent, auto-scaling microservice on Kubernetes.