Engineering Leader - Cloud & AI Platform Architect Founder of https://xitter.cf/t.co/aj9h4UKs4t and https://xitter.cf/t.co/x0g9wqAGo6. Fractional CTO. Actively seeking new opportunities 🚀

Florida
Joined April 2019
Letting AI build your business apps? Give it guardrails. In appmaker.io, sign-in, roles, API keys and encrypted secrets are built in. The AI just configures them. Every change is on the record, and you audit it all in one visual designer before go-live.
1
29
Built an internal tool with AI over a weekend? Nice. Now, who maintains thousands of lines nobody has read? Every appmaker.io app has the same shape and a visual designer. Click a control to see its properties, events and code. Edit it, or ask the AI.
34
Claude Code and Codex are great for bespoke software. But every app they generate rebuilds sign-in, roles and secrets its own way. In appmaker.io, the AI builds on a foundation where that's already done, and you audit the result in a visual designer.
28
Here's the second video ad for AppMaker.io 👀 Also created with Claude Code + Opus 5.5 from a single prompt. Going to test both ads and see which one performs better!
3
3
13,546
First experiment with AI-generated video ads for AppMaker.io 🚀 Built this entire ad with Claude Code + Opus 5.5 using a single prompt. Curious to see how it performs. What do you think?
1
1
9,705
Impressive. Claude Code with Opus 5.5 created two short video ads for me from a single prompt. I'll post them soon and see how they perform!
50
Similar to previous post ... but with a mobile experience. Built a mobile-optimized weather app with AppMaker.io using a single prompt. Responsive layout, touch-friendly UI, and a design made for smaller screens. From prompt to mobile-first app in one shot. 🌤️
1
2
97
From one prompt to a working expense tracker! Built a Pocket Expenses mobile app with AppMaker.io using just a single prompt. Track spending, manage expenses, and keep your budget in check. One prompt. One app. Zero coding. 🚀 #AppMaker #AI
1
72
What can you build with just one prompt? Here's a simple weather app I created with AppMaker.io Clean interface, weather updates, and everything you need at a glance. No setup. No code. #AppMaker #AIApps #NoCode
1
37
From idea to working app, just by chatting with AI. Here's AppMaker's new Build mode in action. Describe your app, watch it take shape, and keep refining it with AI (all while seeing the results live). No setup. No boilerplate. Just build. 🎬👇
38
Big update for AppMaker! 🚀 🌐 appmaker.io We've launched a new website and introduced AI-powered app building. Describe what you want to build, and AppMaker turns it into a working application. You can also switch to the visual designer whenever you want more control.
1
1
51
Code is cheap now, and I think we're just starting to see what that means. Stuff that wasn't worth building before suddenly is. Small teams can build pretty much anything. Sometimes it's easier to build your own tool than to find and buy one.
1
42
"Engineers are no less valuable, but their work is changing." — Anthropic, October 13, 2026.
34
"Our team members still review every change request for the system, but they now examine the intent of the change as opposed to reading every line of code." - from anthropic.com/engineering/re…
35
Another sneak peek at AppMaker's AI Coding Agent appmaker.io 🚀 This time, building a mini LIMS (Laboratory Information Management System) from scratch using AI prompts. Powered by Anthropic Claude Opus 5.5, right inside AppMaker's visual designer. Coming soon!
3
5,267
Building a web app from scratch using just AI prompts. Here's a sneak peek at the new AI Coding Agent in AppMaker, powered by Anthropic Claude Opus 5.5. From prompt to working app, right inside the visual designer. Coming soon to appmaker.io 🚀
1
5,307
Who's using tmux with a coding agent?
48
The key difference between RAG and Agentic RAG is who controls retrieval. In traditional RAG, the application retrieves context before calling the LLM. In Agentic RAG, the LLM decides when to search, what to search for, and whether to search again.
2
1
4
127
For smaller use cases, you may not need RAG at all. If you have just a few documents that fit within the LLM's context window, you can include them directly in the prompt. It's simpler, with no embeddings, vector database, or retrieval pipeline required.
49