{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"Svelte Radio","title":"Benchmarking AI with Stanislav Khromov","html":"<iframe width=\"100%\" height=\"180\" frameborder=\"no\" scrolling=\"no\" seamless src=\"https://share.transistor.fm/e/b519e6cc\"></iframe>","width":"100%","height":180,"duration":4438,"description":"In this episode, Stanislav Khromov joins the Svelte Radio team to discuss his work on Svelte Bench, a\nbenchmarking tool that scientifically measures how well different LLMs understand and write Svelte 5 code.\nThe conversation explores the challenges of AI-assisted coding with Svelte 5, the development of the official\nSvelte MCP (Model Context Protocol) that provides LLMs with documentation and auto-fixing capabilities, and\nthe surprising performance differences between major AI providers. Stanislav also shares insights on working\nwith AI tools, the future of local models, and the economics of AI coding assistants.\nNotes\nStanislav Khromov\nSvelte BenchGitHub\nSvelte MCP\nAnthropic Spending VC Money\nUnpopular Opinions\nKevin: Serverless is overrated\nStanislav: OpenAI has the worst models of the big three\nAntony: Heat Pumps are badTechnology Connections\nTech Ingredients\nPicks\nKevin: Kasai Hibachi Grill (now tested, and can confirm, it is PENG!)\nStanislav: Ball x Pit\nAntony: Saunas By the Sea","thumbnail_url":"https://img.transistorcdn.com/a2spDmSxrcyX_1C736Fgx-G-l9_nFz2rxgwLVukFl-A/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS9zaG93/LzEyODk5LzE1OTc2/Nzg5NDYtYXJ0d29y/ay5qcGc.webp","thumbnail_width":300,"thumbnail_height":300}