Usable AI agents are finally here. From OpenAI’s DeepResearch to xAI’s DeepSearch, we’re seeing the first real push toward autonomous tools that can plan, execute, and complete tasks like research, outreach, and coding with minimal human input. Recently, a new entrant joined the competition — Manus, a general-purpose AI agent that’s getting global attention. Some are calling it the “most sophisticated computer using AI,” while others write it off as just another “wrapper.” So what sets Manus apart under the hood? What does its launch say about the broader debate around GPT “wrappers”? YC’s @garrytan dives into Manus’ “chain-of-thought injection” technique that enables agents to actively update plans, what works particularly well with its UI, and how founders can create enduring value with application-level products.
@ycombinator @ManusAI_HQ Guys please see what Genspark is doing x.com/IBMbotsan/stat…
@ycombinator @ManusAI_HQ Guys please see what Genspark is doing x.com/IBMbotsan/stat…
@ycombinator Woah, cool! Thank you Garry & YC team! Only the beginning for @ManusAI_HQ!
Hi Garry @garrytan , if you have a few minutes, can I ask a small favor - would you try out Genspark Super Agent (a fast and reliable general agent) a bit? User feedback are overwhelmingly positive recently. Your perspective would be incredibly valuable, and I'd really appreciate any feedback you might have.
@ycombinator Exciting to see advancements in AI agents like Manus!
@ycombinator @peakji manus leaked prompt here: github.com/jujumilk3/leak…
@ycombinator @ManusAI_HQ Manus is already old… Genspark AI & Lindy AI are much better and cheaper to use. More tools as well
@ycombinator Chain of AI agents answering questions by building a knowledge graph on podcast transcriptions: x.com/lukaesch/statu…
@ycombinator Chain of AI agents answering questions by building a knowledge graph on podcast transcriptions: x.com/lukaesch/statu…
@ycombinator awesome piece. Makes me wonder about the state of the art in privacy when it comes to sharing our machines and proprietary data with agent/LLM companies. Is there a way to hide the proprietary details while still getting the same performance out of the agent?
@ycombinator I'm cooking open source alternative to Manus at Navigator AI x.com/SohamR_7113/st…
@ycombinator I'm cooking open source alternative to Manus at Navigator AI x.com/SohamR_7113/st…
@ycombinator AI agents are in high demand right now. Every business is looking at ways to automate their process. Manus could be a turning point in the field of AI agents.
@ycombinator Loved Manus. Makes some mistakes, but that's expected. However, the credits exhaust quite fast and hence price could be an issue.
@ycombinator And by the time this was posted Genspark is the new buzzy ai agent
@ycombinator DeepResearch and DeepSearch are interesting but still early. autonomy is coming but hype moves faster than tech. let's see how they handle real tasks.
@ycombinator ImagineAI's focus on ethical AI aligns with this trend, ensuring our agents not only perform tasks but do so responsibly, setting a standard in the industry
@ycombinator @garrytan i know an ai generated Garry tan when i see one
@ycombinator What is the paper that is highlighted in this video?
@ycombinator Hey @ycombinator ! How about AI-generated audio-interview that can be shared across social media in two clicks? Check Aloude
@ycombinator Who is building Supervisor Agents AI ? Like god of agents that can supervise all other agents. It could be a category killer.
@ycombinator Tested 100+ AI agents. Now they handle grunt work, I strategize.
@ycombinator We have models capable of building almost anything we can imagine. The key now is mastering prompt engineering.
@ycombinator @ManusAI_HQ A turning point for productivity, AI agents are no longer experimental, they’re essential. The age of autonomous execution is here.
@ycombinator Proud to be an early investor and supporter of @ManusAI_HQ - this team is going to create a generational business!!
@ycombinator the race isn’t about wrapping GPT anymore it’s about orchestrating agents that can think, plan, and adapt like real teammates @TryAlexAI is built exactly for this not just usable AI but useful from day one
@ycombinator Usable only where correctness doesn't matter. Hopefully that will change soon.
@ycombinator Autonomous agents? Start with pilots—track cohesion metrics. Iterate.
@ycombinator The real breakthrough isn’t smarter chat it’s AI that can step into the work with you, think independently, and actually move things forward. We build with this mindset, & it’s reshaping how we operate. Tools like Manus are pushing this even further. Excited to try it out!
@ycombinator @ManusAI_HQ I'd love to get access to the beta and play around with this. :D
@ycombinator this is next gen stuff! look forward to what we’re able to build with this
@ycombinator Google auth isn’t working any longer in the Manus window when the user takes control and tries to auth into LinkedIn for example.