doozerAI

15 posts

doozerAI banner
doozerAI

doozerAI

@DoozerAI

Communicate and transact business with your customers - on autopilot.

Se unió Mayıs 2023
32 Siguiendo24 Seguidores
doozerAI retuiteado
paulchada
paulchada@paulchada·
🚨 Breaking News: We Just Acquired chaaaat.com! 🎉 While OpenAI splashed $15.5M on chat.com, we went the startup-savvy route and secured chaaaat.com for... well, let's just say we had enough left over for coffee! ☕ 1/3
English
1
2
1
165
doozerAI
doozerAI@DoozerAI·
Our customers just email their digital worker (Hunter) to get things done. Magical! doozer.ai/hunter Built upon the amazing LangChain and our own workflow / decision engine. thanks again @hwchase17
English
0
2
5
2K
doozerAI
doozerAI@DoozerAI·
@jerryjliu0 We’re building virtual AI agents in doozerAI and have created a few YouTube videos. Initially they were temperamental but they are becoming far more reliable and useful. We’re testing your new agents for multiple indexes - working very well!
English
0
1
1
200
Jerry Liu
Jerry Liu@jerryjliu0·
Are you building agents in production? 🤔 There’s been a LOT of hype around agents, and it’s understandable - they’re cool! 🤖 “Do anything” AGI is cool (AutoGPT, SuperAGI, etc.) 🕹️ Agent simulations are cool But there seems to be a big gap between their promise and what people are actually building in more “production-focused” use cases. Had fun chatting about this on the panel with @tuhinone, @mikeknoop, @eladgil during @stripe AI day yesterday: - A lot of enterprise users care about reliability - Agents don't always work - They are just scratching the surface of unlocking value even from “simpler” concepts like retrieval augmentation There’s a spectrum of unconstrained agents to more constrained agents that are good at specific tasks. “Dumber” LLMs require hand-holding and are better suited for more constrained workflows, whereas models like GPT-4/Claude 2 can execute a general purpose reasoning loop (eg ReAct) better - but still may not always work!
 We’ve written about this before: medium.com/llamaindex-blo… I’d love to understand this better. Let me know: 1️⃣ if you’re using agents in enterprise, 2️⃣ what reasoning loop you’re using (ReAct?) and why you trust it 3️⃣ what you’re building with these agents!
English
22
16
143
38.3K
doozerAI
doozerAI@DoozerAI·
Happy Friday!! The first in a series on feature integrations with doozerAI, what better way to kick things off than allowing doozerAI to help you get ahead of your tasks in @todoist before the weekend. Enjoy the weekend 🌞 and see you Monday! 🎉
GIF
English
1
2
7
704
Aayush Mathur
Aayush Mathur@Aayush_Mathur7·
@GregKamradt A network of AI agents, orchestrated and interacting as part of a cohesive team. Consider L6 engineers designing tasks and handing them off to L2 engineers. All the while, a human overseer guides the process, providing approval at each step. AI-assisted workflows is coming.
English
2
2
11
1.3K
Greg Kamradt
Greg Kamradt@GregKamradt·
Pretty remarkable that most companies launch a chat bot I don’t think we’ve seen that horizontal of a product launch so rapidly before What’s next?
English
12
0
28
7.3K
Shaun.AGI
Shaun.AGI@agishaun·
One blog post with one mega thread to understand autonomous agents! This thread is all you need! LLM Powered Autonomous Agents by @lilianweng (leading Applied Research at OpenAI) 🧵👇6 takeaways, 3 components with technics, 4 case studies, 3 challenges
Shaun.AGI tweet media
English
10
56
289
97.9K