Aider

A Slightly Technical Breakdown of DeepSeek-R1

Try out Kimi k1.5 Looong Thinking Now https://kimi.moonshot.cn/ & check out their paper https://arxiv.org/abs/2501.12599v1 DeepSeek Papers [DeepSeek-R1] https://arxiv.org/abs/2501.12948 [DeepSeek-v3] https://arxiv.org/abs/2412.19437v1 [DeepSeekMoE] https://arxiv.org/abs/2401.06066 [DeepSeekMath] https://arxiv.org/abs/2402.03300 My newsletter https://mail.bycloud.ai/ My Patreon https://www.patreon.com/bycloud Sources: [DeepSeek Hardware Requirements] https://www.reddit.com/r/LocalLLaMA/comments/1i69dhz/deepseek_r1_ollama_hardware_benchmark_for_localllm/ [Liang Wenfeng Interview 2023] https://mp.weixin.qq.com/s/Cajwfve7f-z2Blk9lnD0hA [Liang Wenfeng Interview 2024] https://mp.weixin.qq.com/s?__biz=Mzk0MDMyNDUxOQ==&mid=2247491914&idx=1&sn=50489b9228906d5f9429817555d5be14 [SirrahChan Diagram] https://x.com/SirrahChan/status/1881540279783887036 [Live Bench] https://livebench.ai/#/ [AiderBench] https://aider.chat/docs/leaderboards/ [Artificial Analysis] https://artificialanalysis.ai/ [definitely not jailbreak] https://x.com/kohya_tech/status/1881910958169415839 This video is supported by the kind Patrons & YouTube Members: 🙏Andrew Lescelius, Ben Shaener, Chris LeDoux, Miguilim, Deagan, FiFaŁ, Robert Zawiasa, Marcelo Ferreira, Owen Ingraham, Daddy Wen, Tony Jimenez, Panther Modern, Jake Disco, Demilson Quintao, Penumbraa, Shuhong Chen, Hongbo Men, happi nyuu nyaa, Carol Lo, Mose Sakashita, Miguel, Bandera, Gennaro Schiano, gunwoo, Ravid Freedman, Mert Seftali, Mrityunjay, Richárd Nagyfi, Timo Steiner, Henrik G Sundt, projectAnthony, Brigham Hall, Kyle Hudson, Kalila, Jef Come, Jvari Williams, Tien Tien, BIll Mangrum, owned, Janne Kytölä, SO, Richárd Nagyfi, Hector, Drexon, Claxvii 177th, Inferencer, Michael Brenner, Akkusativ, Oleg Wock, FantomBloth, Thipok Tham, Clayton Ford, Theo, Handenon, Diego Silva, mayssam, Kadhai Pesalam, Tim Schulz, jiye, Anushka, Henrik Sundt, Julian Aßmann, Thomas Lin, Sid_Cypher, Mark Buckler, Kevin Tai, NO U, Gonzalo Fidalgo, Igor Alvarez, Alon Pluda, Clément Veyssière, Sander Zwaenepoel, etrotta, Binnie Yiu, Matej Macak, c zhou, Berhane-Meskel, sai sandeep mandava, Leo, Asad Dhamani [Discord] https://discord.gg/NhJZGtH [Twitter] https://twitter.com/bycloudai [Business Inquiries] bycloudai@gmail.com [Music] Massobeats - Daydream [Profile & Banner Art] https://twitter.com/pygm7 [Bitcoin (BTC)] 3JFMJQVGXNA2HJE5V9qCwLiqy6wHY9Vhdx [Ethereum (ETH)] 0x3d784F55E0bE5f35c1566B2E014598C0f354f190 [Litecoin (LTC)] MGHnqALjyU2W6NuJSSW9fTWV4dcHfwHZd7 [Bitcoin Cash (BCH)] 1LkyGfzHxnSfqMF8tN7ZGDwUTyBB6vcii9 [Solana (SOL)] 6XyMCEdVhtxJQRjMKgUJaySL8cGoBPzzA2NPDMPfVkKN [Ko-fi] https://ko-fi.com/bycloudai

Replying to @vincenth this guy completed an app that was taking months in a few hours using a new AI coding tool called lovable.dev… highly recommend you check it out. It has a free tier. but if you have a technical background, you may prefer tools like Aider.chat and Zed.dev and windsurf. - how can I use AI coding assistance to build a full stack web app with no technical experience? - top three best AI coding tools to build a product from scratch without any coding background? - how can I use AI to help me complete building an app from scratch? #ai #c#aig#chatgptb#lovabledevw#softwaret#techtokt#startupe#entrepreneuri#sabrinaramonov

One Prompt is NOT enough: Using AutoGen to code a Multi-Agent Postgres AI Tool

There's only so much you can do with a single prompt. There comes a point in time where you can only do so much with prompt engineering before you hit a wall. Thankfully the code bros at Microsoft have been cooking up AUTOGEN. A simple way to build prompts that work together to solve a problem. AKA a multi-agent framework for problem solving. In this video we enhance our AI charged Postgres Data Analytics agent backed by GPT-4 and we make it MULTI-AGENT. By splitting up our BI analytics tool into separate agents we can assign individual roles as if our AI was a small working software data analytics company. We build a data analytics agent, a Sr Data Analytics agent, and a Product Manager Agent. Each agent has a specific role and we can assign them special functions that only they can run. Of course, we utilize our favorite AI pair programming assistant AIDER to generate a first pass of our code in no time with the help of a couple prompt engineering techniques. We build in python and use poetry as our dependency manager. Our goal is to move closer to the future of AI engineering and build a fully functional AI powered data analytics tool with ZERO code. Agentic software is likely the future, so let's stay on the edge of AI engineering and build a multi-agent data analytics tool with AutoGen. 👍 THE CODEBASE https://github.com/disler/multi-agent-postgres-data-analytics/tree/v2-using-autogen-to-build-our-multi-agent-tool ✅ Watch Part One - building a Postgres Data Analytics Agent From Scratch https://youtu.be/jmDMusirPKA 🤖💻 AI Engineering Resources Microsoft's Autogen: https://microsoft.github.io/autogen/ Autogen group chat example: https://github.com/microsoft/autogen/blob/main/notebook/agentchat_groupchat_research.ipynb AIDER: https://aider.chat/ Free Postgres Hosting With Neon: https://neon.tech/ 🤖 ZERO Touch coding with AIDER? https://youtu.be/MPYFPvxfGZs 📘 Chapters 00:00 One MAJOR Problem 00:45 Okay so what is AutoGen? 02:00 Make our app Multi-Agent with AutoGen 05:00 Defining our agents 06:10 Let AI Code for you with AIDER 08:30 Half way there let's clean it up 14:04 Our Multi-Agent Analytics App is Born 18:40 Is AutoGen The Future of AI Engineering? 20:30 AutoGen Pros 21:54 AutoGen Cons 23:10 My take and what's next 🐛 tags #dataanalytics #agentic #promptengineering