the watcher

VCs used to worry frontier labs will kill applications. Luckily, with open source models, suddenly those application companies will be just fine, with their abilities to collect context and route models. Companies like replit, lovable, genspark used to have negative margins and terrible unit economics can now become profitable and competitive again. Expect to see more investments in applications, previously known as "wrapper" companies.

Been thinking about this: 1. Spotify have not publicly talked about engineering in quite some time 2. Many (most?) software engineers are paid Spotify customers, mildly annoyed at the product 3. So when they saw Claude + Spotify they immediately started to air grievances

Price per token != cost per task

“Loop engineering” is a hot buzzphrase after mentions of it by Boris Cherny (Claude Code’s creator) and Peter Steinberger (OpenClaw's creator) went viral on social media. Loops are now a key part of how we get AI agents to iterate at length to build software. In this letter, I’d like to share my 3 key loops, shown in the image below, for building 0-to-1 products. These loops guide not just how I build software, but also how I decide what software to build. Agentic coding loop: Given a product specification and optionally a set of evals (that is, a dataset against which to measure performance), we can have an AI agent write code, test its work, and keep iterating until the code is bug-free and meets its specification. This idea of closing the loop took off around the end of last year, and it has been a game changer in enabling coding agents to work longer productively without human intervention. For example, over the weekend, I was building an app for my daughter to practice typing, and my coding agent could easily work for around an hour, using a web browser to check what it had built multiple times before getting back to me, without needing my intervention. The engineering loop executes quickly. Every few minutes, the coding agent might build and test a new version of the software. I hear frequently from developers who are finding new ways to engineer more effective engineering loops. This is an active area of invention! Developer feedback loop: In this loop, a developer examines the current product and steers the coding agent to improve it. Last year, a lot of developers (including me) were acting as the QA (quality assurance) function for our coding agents, manually finding bugs and then asking the agent to fix them. But with coding agents much more able to test their own code, the amount of time we need to spend on this function has decreased significantly. This allows us to make higher-level product decisions, such as what key features to offer, where the UI needs improvement, and so on. The developer-feedback loop operates over time intervals between tens of minutes and hours — that's how frequently a developer might review a product and give feedback. In the case of the typing app, I changed my mind a few times about the visual design, what cat costumes she can unlock as she learns (she loves cats), and the user flow for a grown-up to log in and steer the child's learning experience. When a developer has a clear vision for what to build, it is still a lot of work to translate that vision into a specification for a coding agent to implement. Further, after the developer has seen an implementation, they might update (or perhaps clarify) the spec to steer it toward what they want. If you find that the system repeatedly runs into certain problems, building a set of evals for the agent becomes useful. AI-native teams are increasingly using AI to help shape product direction, for example, automating the gathering and analysis of usage data, summarizing written and verbal customer feedback, or carrying out competitive analysis. However, for pretty much all the products I’m involved in, I see humans as having a significant context advantage over current AI systems — we know a lot more than the AI system about the users and the context the product has to operate in — and thus humans play a critical role. Many people describe this human contribution as “taste,” but I prefer to think of it as humans having a context advantage, since that gives us a clearer path to helping AI systems get better. This also speaks to why this step can’t be automated: So long as the human knows something the AI does not, human-in-the-loop is needed to to inject that knowledge into the system. External feedback loop: This includes a wide range of tactics like asking a few friends for feedback, launching to alpha testers, or putting the code into production with A/B testing. These tactics are usually slow, rarely taking less than hours and sometimes taking days or even weeks. This data informs the developer vision, which in turn continues to drive the detailed product spec, which in turn drives the coding agent. With coding agents speeding up software development, more engineers are starting to play a partial product management role. For many engineers who are growing into this role, the hardest part is shaping the product vision and striking a balance between building (bridging the gap between vision and spec) and getting user feedback to evolve the vision. It is important to do both! I will write more about how to do this in future posts, but for now, I find it encouraging that engineers are playing an expanded role (just as product managers and designers now do more engineering). [Original text: The Batch]

Sonnet 5 initial impressions: >much more optimized for autonomous execution >better at planning, tool use and implementing complete tasks >more likely to do too much when you only want advice or a small change >sometimes ignores narrow constraints and jumps straight into implementation >stronger as an autonomous agent, but potentially worse as a collaborative coding assistant The paradox: ❌Optimizing models to replace the developer may make them worse at working alongside one. that's why you should stick to GLM-5.2, Kimi or smaller models for daily assistance... they may be less capable overall, but they’re more predictable and more likely to do exactly what you asked. Tip: Use Sonnet 5 for clearly scoped end-to-end tasks. For pair programming, precise edits and human-in-the-loop workflows, a more obedient model may actually be the better choice.

การมาล่าเงินพ่ออีลอน หลักๆไม่มีอะไรมาก แค่ฟอลติ๊กฟ้า ให้เยอะเข้าไว้ อย่าไปอัลฟอล อย่าไปมิ้ว มันต้องเปิดใจรับในความต่างของกันและกัน ฟอลแล้วมิ้วทำไปเพื่ออะไรก็ในเมื่อเราต้องฟอลติ๊กฟ้าเยอะๆ เพื่อให้มีปฏิสัมพันธ์และการมองเห็นกันมากยิ่งขึ้นไม่ใช่หรือ ค่อนข้างจะงงเวลาที่ฟอลกันแล้วมิ้ว ** ต้องทวิตเอง เป็นโพสต์บ้างคลิปบ้างรูปบ้าง อย่าแค่บ่นว่าทวีตอะไรไปก็ไม่มีใครเห็น เอาแต่ทวิตบ่น เอาแต่นอยด์ ** ไปคอมเม้นท์ให้หลากหลาย ไม่ใช่เกาะแต่แอคใหญ่อย่างเดียว ** รีไปเถอะ แล้วกดใจด้วย ไม่ว่าจะของใครแอคที่เราฟอล หรือแม้แต่ไม่ได้ฟอลกันก็ทำได้ ** ทั้งรี ทั้งกดใจ ทั้งคอมเม้นท์ ถ้าทำได้ยิ่งดี แต่ห้ามทำเหมือนบอท 📌 มีหลายแอคมาเล่นกับแอคเรา เราเลยอยากเข้าไปรีไปดันให้ แต่เข้าไปกลับพบว่า เค้าไม่ทวิตอะไรเลย นอกจากรีทวิต กับคอมเม้นท์ในทวิตคนอื่น อย่าลืมนะเราไม่มีทางรู้หรอกว่าทวิตไหนของเรามันจะแมส เพราะฉะนั้นทวิตไปเลย อย่าไปกังวลว่าจะมีใครมาตอบโต้รีทวิตให้เราหรือไม่ 🤟🏻 📌 สุดท้าย อย่าลืมว่า ถ้าเราซัพพอร์ตเขา เขาก็จะซัพพอร์ตเรากลับเช่นกัน 📌สุดท้ายนี้ ขอให้ทุกคนที่กำลังสร้างรายได้ในx สำเร็จตามเป้าหมายที่ตั้งไว้ 🙌🏻✌🏻🍀 #ติ๊กฟ้า

We’ve received notice that the Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5. We'll begin restoring access tomorrow, and will share an update soon. We’re grateful to our users for their patience, and to everyone who worked with us on redeploying the models.

we launched code interpreters for deep agents last month. Basic idea is to let agents plan, delegate, and organize context using code instead of chained tool calls Code interpreters don't need a sandbox, but we still need a way to securely run that code! (and running untrusted code is a famously hard problem) Here's the writeup on how we're looking to do just that: