Mingwei Shen

I can’t even do that on flat ground with running shoes in normal conditions

@YiTayML But you can cache your input tokens? your decoder (story telling) is likely probabilistic and will be different each time but your inference speed is faster.

my 2 year old keeps requesting us to read her the same storybook again and again. thats cute but repeating tokens is bad should I tell her that?

Quick question for Open AI app developers - how do you test before publishing updates? Manual testing, scripts, more formal? Exploring if automated testing and monitoring is useful vs today’s options. Curious about your workflow and pain points. #OpenAI #BuildInPublic

@johnrushx Have you tried curating your own “memory” and storing as a RAG? Basically you can prompt the agent to refer new context to your RAG. You basically need a way to transform context with pre-defined, customizable weights. RAG is just a tool to achieve that. there will be better ones

I’m feeling so pessimistic about AI today. After building dozens of AI agents for many years, here is what I got to say: The biggest challenge in building AI agents is agent’s long term memory. LLMs with huge context windows are pretty much scams, cuz they compress the text under the hood making it completely unreliable for any serious task (the error accumulation and forward propagation when run in an agentic loop ). Existing LLMs would be capable of replacing humans if only the memory problem was really solved. But..nobody has come even close to solving this yet, and it may take a decade until this is solved. I won’t even be surprised if it turns out that our general intelligence comes from our insanely amazing memory engine in our brains coupled with the logical apparatus. Right now the LLM can only apply logic, but it can’t memorize things as our brain does, therefore we’re very far from AGi or even a basic 100% autonomous ai agent. Unfortunately :( But, I have an idea to solve this at least for business ai agents. I’ll present my solution soon in my next ai agents release.

Just read an article that only 5% of users of ChatGPT are paying users. While it’s costly as a business, they are also subsidizing the 95% of users who would not have access to the same level of ai tools otherwise. Thankful to @OpenAI for enabling access to the masses!

@ElliotEvNo I hope you live your dream. Fearless builders make ideas reality. Regardless of what happens, I’m sure you’ll have a great experience!

i'm leaving Lovable at 16 y/o to build AGI. the bottleneck isn't smarter models, it's the tools and context around them. more soon.

@frantzfries I’m hopeful this becomes the next Apple Store - democratizing ways for builders to monetize ideas lowers barriers to entry

if I was in consumer or looking to build for consumers I would drop everything and go all in for a month on this might be a small chance it becomes the next App Store but imagine if you're one of the 3 fitness/dating/sudoku/whatever apps when it does

@jacobrodri_ Curious about your findings - especially as you get paying customers and add more features post launch. Followed and looking forward to your updates!

Which AI app builder would you say is the best right now? Planning to try one soon

@JonhernandezIA I actually like that I can see what’s being built. I value simplicity of the lovable interface, and I value the ability to visualize things when I want to go deeper on something.

The future doesn't look like this to me. I have the feeling we should have skipped that and gone directly to a natural language builder.. Why is this not a lovable style interface that THEN let's you tweak the nodes if something doesn't work??

@GergelyOrosz I think it’s amazing how fast things are being shipped these days

I really don't think most people appreciate the difference between shipping something in 6 weeks, beating the global competition (first to offer an agent workflow builder!), then fixing up small issues in 6 more weeks. Vs building for 6 months, and no one cares by launch...

Love this concept - fostering the next generation of all-in-one egalitarian entrepreneurs!

@paulg @sama I’m thinking “good actor” volumes may eventually dwarf “bad actors”. a few personas to grow specific brands, one personal and others for specific causes/ideas. agents to generate content and interact with community => reminds me that 99+% of energy in the universe is dark matter

@sama I've noticed more and more in my replies. And not just from fake accounts run by groups and countries that want to influence public opinion. There also seem to be a lot of individual would-be influencers using AI-generated replies.

i never took the dead internet theory that seriously but it seems like there are really a lot of LLM-run twitter accounts now

@BennettBuhner lol this is hilarious

WHY IS THE NEW QWEN3 MODEL SO ZESTY WHAT IS THIS???

@vitrupo @huggingface From a builder point of view, has open source won in the past? Incentive structure wise, it’s not clear that open source benefits the creators. Even meta with its fat margin threw in the towel. I would love to understand how open source can be competitive.

Jack Dorsey says AI must be permissionless because constraint kills innovation. Five CEOs shouldn't dictate what brings humanity forward. Open source is the answer. To protect ourselves, we have to race ahead. Eliminating single points of failure before they become civilization's choke points.

@jeffwsurf Glad that Windsurf is still alive and well, and that its builders are taken care of. Looking forward to future improvements to Windsurf! Or is it Devinsurf now?

To put it mildly, the past week at Windsurf has been crazy. There have been a lot of different rumors and reports, so I want to share a transparent account of how it actually went down. Before I start, I just want to say that Varun and Douglas were great founders and this company meant a lot to them, and it should be acknowledged that this whole situation must have been difficult for them as well. One week ago, last Friday, I walked into the office for our all-hands, where our ~250 people were expecting to hear that we were getting acquired by OpenAI. By that time, I had already learned what was really about to happen and had broken the news to Graham, the new President, and Kevin, the new CTO. You can imagine the shock when the team found out. It was my job to explain to the company our path forward as a company. In my view, there were a few options: try to raise more money (VCs were offering), try to sell the company (we had interest from multiple parties), try to distribute (although there were liabilities to wind down), or try to keep running the company. Although we had lost some great people and taken a serious blow to morale, we still had all of our IP, product, and strong talent including an excellent GTM machine - the core components of the company were still there. The mood was very bleak. Some people were upset about financial outcomes or colleagues leaving, while others were worried about the future. A few were in tears, and the Q&A had been understandably hostile. In distress, people asked if we could distribute the cash immediately, but we also needed it to pay bills and keep the product working for customers. After trying to steady the team as much as we could, Graham and I spent the rest of the evening on calls, trying to identify every available option. Out of the blue, we got a text and email, from Scott and Russell, with the message, “Chat?” It was around 5:30pm, exactly this time last week. I immediately called Graham and told him I thought this combination made sense. Cognition had been the one team our people had respected the most. While they had overinvested in engineering, they had frankly underinvested in GTM and Marketing, and our teams in those functions are nothing short of world class. On the other hand, we now were missing a Core Engineering team, and there’s no better group of AI engineers than the lineup Cognition has assembled. Then, there was the product logic. Devin would benefit from a foreground synchronous agent, while we needed a remote asynchronous agent. The teams and products together would be able to create an unrivaled end to end platform. We took the Cognition approach very seriously from the start and launched right into negotiations. Scott and his team moved fast, and while the timeline was exploding with memes and commentary between Friday and Monday, Scott and Russell spent that time in our office, working tirelessly to get a deal done in record time. Saturday rolled in, and I brought Kevin to the office. At this point, we were still having 1:1s with our enterprise engineers to retain them, and at the same time, we were trying to get more information to diligence Cognition while reviewing other potential partners. Saturday, I was still getting inbound interest from potential acquirers, including one we had all looked up to for a long time. But by then, Scott was already in our conference room with physical papers to sign and was handing me a pen. We had already made up our minds that there was no better partner than Cognition, even with other excellent and impressive companies interested. We quickly brought in lawyers to review the LOI and signed that day, a little over 24 hours after Scott’s cold outreach. Saturday had been spent understanding each other’s businesses, and when the sun came up Sunday morning while we were in the office, we prepared to move on to getting the deal finalized. The Windsurf team had been through too many twists and turns, so both Scott and I felt that the next transition for them had to truly be final. The other priority during this process, which Scott and I aligned on and was one of the things that helped me realize he was the right partner, was that we needed to take care of all Windsurf employees. Their work and talent had gotten us to this point; they deserved to be paid for that, and we wanted to give them a better price than any of the previous scenarios. That resulted in a key part of the deal: structuring it to give a payout to every employee, to waive all cliffs, and to accelerate all vesting for Windsurf equity. On Sunday, an army of lawyers from both sides descended on the office, having been challenged by Scott and Russell to get this deal finalized within 24 hours. We ate and slept (or at least tried to grab quick naps between discussions) in the office through the weekend to get it done. Our teams and lawyers stayed up all night Sunday working out the final details. On Monday morning, we went over every detail again, got board approvals, and the documents were ready to sign. One of our lawyers said it was one of the fastest deals they’d seen. Monday 9:30am, we signed our definitive agreement for Cognition to acquire Windsurf. Scott had given a heads up to Cognition employees, and we had another all-hands scheduled for Windsurf employees at 10am that morning. After the traumatic announcement on Friday, Scott and Russell wanted us to be able to open Monday with good news and we wanted to handle this new announcement in a much better way for employees. In fact Russell took a red eye the night before and somehow made it to the Austin office just in time. We held Monday’s all-hands in the same room as Friday’s, and this time Scott was by my side at the front of the room. It was a blur, but the highlights I’ll always remember were getting to tell employees what we had negotiated for them (“We’ve decided to give you 1 year of vest, OH, and years 2, 3, and 4 as well!”) and Scott saying, “A founder goes down with the ship.” The applause from our people seemed to last forever, and I was on the verge of tears myself. Now comes the real work. We are officially one company, operating as two entities, but with much to do both internally and externally. We have work to do on both our team and on our product, to realize our shared ambitions. It was a wild ride this week, and now the story has been told. We are excited to move on to the next chapter. We’re putting our heads back down to focus on building the future of AI together.

Worth a read on how seasoned devs can leverage GenAI to increase velocity while integrating best practice frameworks in acceptance criteria, design, implementation/testing to version control and document key decisions.

Worth a read to get insight into what it is like to work at open ai #footnote-fnref-1" target="_blank" rel="nofollow noopener">calv.info/openai-reflect…

@swyx Ryan Kaji is the standout winner for me. Kid’s amazing

@swyx Wow those are really big numbers. Now I have a list of people to go figure out what they actually do :)

one way to view the $75m/yr MSL comp packages is that it will materially change the aspirations of our next generation. people who lamented that our kids who only aspire to grow up to be youtubers and tiktokers now have new heroes to point to, whose work hopefully creates agents that free up time rather than consume it.

Kiro has been awesome so far! Claude 4 is super powerful and gave a nice upgrade over windsurf. The specs, testing and agentic features are super interesting. Will be playing with this for the next few weeks!