Supabase has built a $5B platform by turning high-quality docs into a massive community distribution loop. Here's how they structure their Changelog page to keep 4M+ developers shipping.

Many devtools don’t fail because the tech is bad. They fail because nobody actually builds with them. I’ve seen tools with amazing docs. At some point you stop reading and think, “okay but what does this actually do?” If devs can’t see real integrations, they won’t try your tool. Adoption starts with proof. What do you think? Drop me a DM → Get a 10-min Pilot Fit Check for your DevTool

Before crashing last night, I wrote out detailed docs for a new feature combining a few tools. I went to bed and told the model to just approve whatever it needed to and push through so I could check it out in the morning. The output it gave me has me feeling overwhelmed. Everything I’ve envisioned for the last 3 years when I started building this company is at my fingertips. I feel terrified, anxious, excited, and my heart has been pounding non-stop. If you’re a technical team playing in an industry you have not lived and breathed, I think the next 12 months are going to be painful. Congrats on your last raise. A non-technical SME can now do everything you can, but better. For the first time in software, the scale has shifted to the subject matter experts. The ones with deep industry knowledge. The ones who actually have a passion for the problems they’re solving because they’ve lived the problems for years. I can’t wait to show the title industry what we’re building at Zitles

At Jams, we’ve been experimenting with a powerful new stack: Bubble + Convex + Clerk. 🚀 The results have been incredible, we’re seeing up to 95% in cost savings. By bypassing Bubble's servers entirely, our Workload Unit (WU) consumption is down to the absolute minimum (just essential page loads). All our mutations, queries, and actions now live completely in Convex. Many developers make the expensive mistake of connecting to an external database using the API Connector. This effectively doubles your costs: you burn through WUs and pay for the external DB service. After seeing phenomenal results with our internal apps, we've transitioned to using this stack for our real-world projects. We have the entire process documented and a team fully equipped to handle this new standard of building. At the end of the day, this is a massive win for our clients. They get: ✅ A genuinely scalable application ✅ Blazing fast, real-time performance capable of handling millions of queries and mutations ✅ An architecture that makes future migration to traditional code incredibly easy

The modern enterprise needs more than just another piece of software. It needs a fundamental rethink of how financial data is structured and scaled. We’re proud to have Crystal Huang and the team at GV (Google Ventures) as partners in this mission. GV has a long history of backing the most ambitious technical founders, and their support is a testament to the scale of the problem DualEntry is solving.

"In five years, there’s no reason why every company shouldn’t be using DualEntry.” ERP is a massive category. Earning that kind of future requires focus and follow-through. From day one, the DualEntry team has shown the ambition, velocity, and clarity to meet that challenge. Hear GV General Partner Crystal Huang’s share why we believe in DualEntry’s vision.

We've committed to a zero bug policy at Sourcebot, and we got there using Linear. We believe this the only way to build generational software tools. This idea started when we switched to using Linear a few months ago. Switching from our previous tool to Linear was night and day. Everything just worked! When we submitted a support request they would reply within 24 hours! It made us realize how accustomed we are to using buggy software everywhere. We want developers to have the same experience when they use Sourcebot. Linear is committed to a zero bug policy, so we decided to commit to one as well. The great thing with using Linear is that it supports a zero bug policy natively. I wrote a quick blog post describing how we follow this policy with Linear's GitHub integration, SLA, and dashboard features (link below).

The last few weeks have been some of our best at Plain. This week, we're going further. We've shipped Workflows, our MCP server, new AI features, and countless improvements. Every release is unlocking more of what we set out to build. Starting tomorrow, we're rolling out four new features we've been working on for a while. All focused on making sure support doesn't become the bottleneck as your company scales. One post a day. We're making Plain the platform teams don't outgrow.

The Redgate data summit is 'where database professionals come together to gain the insights needed to solve complex challenges of database management across the Database DevOps lifecycle, and realise value and unlock growth through AI'. You can also hear James, along with other presenters talk on "Tackling AI in the Real World: Complexity, Data Quality, and Rapid Change" and we'll have our managed service team on hand for questions! It's free to attend, and a great opportunity to hear from industry leaders and connect on relevant topics.