Gagan Bansal
@bansalg_Building #AutoGen @Microsoft | Alum of @uwcse, @iitdelhi | Follow @pyautogen for updates on #AutoGen
Similar User
@tongshuangwu
@d19fe8
@fabulousQian
@huashen218
@CaoHancheng
@merrierm
@QVeraLiao
@HelenasResearch
@lxieyang
@dsweld
@joon_s_pk
@TobyJLi
@dakuowang
@bryanhaoenwang
@SaleemaAmershi
Excited to finally release Magentic-One! The thing I love about this multi-agent team is that the same implementation achieves very strong performance across three challenging agentic benchmarks. If you are someone working on agentic systems, you know how challenging this can…
📢Introducing Magentic-One, a generalist 5-agent multi-agent system for solving open-ended web- and file-based tasks. 🤖🤖🤖🤖🤖 Magentic-One represents a significant step towards agents that can complete tasks that people encounter in their daily lives and can achieve strong…
5/ Official statement from @pyautogen about AutoGen forks and clones and the current pyautogen package. Very interesting. x.com/pyautogen/stat…
To our collaborators & community: We’ve seen questions about AutoGen forks/clones vs. the official project. Here's a summary of the latest. Please share with others. - The official repo is github.com/microsoft/auto…. - We're actively working on AutoGen v0.2, with v0.4 innovations…
👀
📢As part of our ongoing work to improve AutoGen, we have been rewriting AutoGen Studio. It is still early but here are some of the new features enabled by the AutoGen 0.4 preview API - Real-time agent updates streaming to the frontend - Run control: You can now stop agents…
If each step of an ai agent is 95% accurate. None of the 30 step work flows will work. Going from 95-> 99.9 is a similar last mile problem as with self driving cars. Easy to hack up a prototype. Hard to make it work reliably at scale.
To all our collaborators and community members, There have been a number of asks to clarify and distinguish between various AutoGen forks and clones. Please check out our recent statement below 👇 #AutoGen #OpenSourceAgents
"The key thing that Microsoft is trying to go for here is to basically have a multi agent architecture where it can generalize to a whole bunch of different tasks without having to have fine grained, step by step instructions." - @Sam_Witteveen (YouTube) Video:…
"It's better than almost every other AI agent that I have ever seen. I may just keep it above every other one including the computer use agent..." -- AICodeKing (YouTube) Link to the code in the bio. #AutoGen #MagenticOne #OpenSourceAgents
Screen parsing continues to become popular!
Announcing llama-ocr – a free + open source OCR tool! It takes documents (images for now) & outputs markdown, and does really well for complex receipts, PDFs with tables/charts, ect... Powered by Llama 3.2 vision on @togethercompute & available on npm today!
Fantastic work from fantastic team @kylelostat @dsweld Is there an easy to use demo script to go from set of papers to table?
✅dataset, ✅framework, ✅eval for taking a bunch of research papers and generating a structured table for lit review 📋 fun project that merges info extraction, QA, and summarization 😊
Great to see reception of #AutoGen Magentic-One!
微软推出了Magentic-One，这是一个协调多个代理执行现实任务的代理框架。 该代理可以编写代码、操作浏览器，甚至从餐馆订餐。 AI代理时代即将到来。 x.com/rowancheung/st…
So much of this was overshadowed this week on X from the election. My favorite was the Microsoft Agent!
Microsoft introduced Magentic-One, an agent framework that coordinates multiple agents to tackle real-world tasks. The agent can write code, operate a browser, and even order food from a restaurant. The AI agent era is coming x.com/rowancheung/st…
Microsoft introduced an Agent framework this week, and it went completely unnoticed. The agent team can browse the internet, internal files, execute code, and more. In this example, the Microsoft team made the agent autonomously order a sandwich. First, an agent: A) Plans -…
Microsoft introduced an Agent framework this week, and it went completely unnoticed. The agent team can browse the internet, internal files, execute code, and more. In this example, the Microsoft team made the agent autonomously order a sandwich. First, an agent: A) Plans -…
Excited to release our agent team Magentic-One! Magentic-One can browse the web, files, write and execute code & supports a human-in-the-loop. Built by @MSFTResearch AI Frontiers with @pyautogen aka.ms/magentic-one-b… My favorite task:
Magentic-One is a multi-agent platform for building teams of LLMs that solve complex tasks. Five core agents have special abilities, including an Orchestrator that plans, tracks progress, notes when stuck, and works to re-plan if necessary. @MSFTResearch microsoft.com/en-us/research…
📢Introducing Magentic-One, a generalist 5-agent multi-agent system for solving open-ended web- and file-based tasks. 🤖🤖🤖🤖🤖 Magentic-One represents a significant step towards agents that can complete tasks that people encounter in their daily lives and can achieve strong…
Microsoft introduces Magentic-One, a new generalist multi-agent system designed to handle complex web and file-based tasks. It uses an Orchestrator agent that directs four specialized agents: WebSurfer for browser operations, FileSurfer for file management, Coder for programming…
1/ @pyautogen announced the preview of the new AutoGen architecture (coming in v0.4)! Can’t wait to check this out. Great work @bansalg_ @ekzhu + team x.com/pyautogen/stat…
We are excited to announce a preview of the new architecture of AutoGen (coming in v0.4). To learn more, see Blog: microsoft.github.io/autogen/blog/2… Pull request: github.com/microsoft/auto… Come help us shape the future of AutoGen!
Yep. I have been waiting for my green card for like the last 3 years. Still haven’t gotten it. People mostly have no idea when they talk about immigration.
One thing I find about the immigrants dialogue is how few of the people making it seem to know how hard it is to actually immigrate to this country. I've been an immigrant 3x now and US was by far the longest and hardest it's been.
Analysis of o1-preview on medical benchmarks, comparing o1's native inference-time powers to prior work on Medprompt, a different approach to inference-time analysis. Assessment: o1 reaches new heights--and shows great promise for medical applications. tinyurl.com/9yspcpm6
United States Trends
- 1. Jake Paul 1,04 Mn posts
- 2. #Arcane 242 B posts
- 3. Good Saturday 28,7 B posts
- 4. Jayce 56,3 B posts
- 5. #SaturdayVibes 3.509 posts
- 6. #saturdaymorning 2.319 posts
- 7. Serrano 242 B posts
- 8. Pence 83,4 B posts
- 9. #SaturdayMood 2.054 posts
- 10. #PlutoSeriesEP5 177 B posts
- 11. AioonMay Limerence 141 B posts
- 12. IT'S GAMEDAY 2.774 posts
- 13. Vander 18,6 B posts
- 14. $WOOPER N/A
- 15. Caturday 8.295 posts
- 16. maddie 22,6 B posts
- 17. Fetterman 38,1 B posts
- 18. WOOP WOOP 1.516 posts
- 19. $XRP 88 B posts
- 20. John Oliver 15,2 B posts
Who to follow
-
Sherry Tongshuang Wu
@tongshuangwu -
Ken Holstein
@d19fe8 -
Qian Yang
@fabulousQian -
Hua Shen✨
@huashen218 -
Hancheng Cao
@CaoHancheng -
Meredith Ringel Morris
@merrierm -
Vera Liao
@QVeraLiao -
Helena Vasconcelos' Research
@HelenasResearch -
Michael Xieyang Liu
@lxieyang -
Daniel Weld
@dsweld -
Joon Sung Park
@joon_s_pk -
Toby J. Li😺 (he/him)
@TobyJLi -
Dakuo Wang@CSCW&EMNLP, Hiring PhDs/PostDoc
@dakuowang -
Bryan Wang
@bryanhaoenwang -
Saleema Amershi
@SaleemaAmershi
Something went wrong.
Something went wrong.