A masterclass in controlled chaos 😛

AI tools multiply, and so do config files. “This has all given birth to a plethora of rule files that we need to manage.” Vibe-rules CLI discovers and syncs them so you can code, not babysit .yaml. Details ➜ ainativedev.co/00n #AINativeDev #DevTools #AIDevelopment #CLI

From assembly → C → Python → Chat. Does Lovable 2.0 belong in the next abstraction layer? Explore this exciting news with Bap -> ainativedev.co/fzn #AINativeDev #DevTools #Lovable

Interesting to see Lovable balancing abstraction with more hands-on control. Will be watching how Dev Mode evolves.

I like the idea of parallelizing Claude sessions with different specs. It's a great way to explore multiple implementations at once.

History repeats: Visual Basic empowered non-coders decades ago. Lovable 2.0 now does the same with AI. Can it truly democratize app-building, or will dev oversight remain essential? Explore this exciting news with Bap -> ainativedev.co/fzn #AINativeDev #DevTools

The “big brain vs open ecosystem” point resonated. Central control can be efficient, but brittle…

GitHub released its own MCP server. You can now use Copilot to: - Find files across a repo - Open and close issues - Query repo state via natural language With Microsoft just open sourcing Copilot, this isn’t just about code-gen. It’s about delivering richer context, greater

One of the most interesting announcements at I/O got just a few seconds of airtime: Gemini Diffusion. A diffusion-based language model that: • Generates in parallel, not left-to-right • Hits ~1500 tokens/sec • Can self-correct mid-generation A potential shift in how AI

Sourcegraph's Amp is now available to all. Built for agentic coding, Amp is designed to handle complex, multi-step development workflows. Thorsten Ball notes: “Amp now writes 70–80% of the code I commit.” Dion Almaer comments on this release: ainativedev.co/jnw

“2025 is the year of agents,” says Greg Brockman. In line with that, OpenAI has reintroduced a revamped version of Codex with Codex Agent. Dion Almaer dives into the details and shares his thoughts: ainativedev.co/q3i #AINativeDevelopment #AIForDevelopers #CLI #Codex #ClaudeCode

On that same thread, I'd also recommend the Firecrawl and Obsidian MCP.

JVM devs, this one’s for you: Rod Johnson just dropped an OSS agent framework. JVM-native, Kotlin-based, and focused on agentic flows mixing LLM-prompted interactions with code and domain models. Dion Almaer got hands-on and shared his thoughts here → ainativedev.co/1i9

This is super exciting stuff.

I use playwright-mcp to drive the browser, but why not go directly to CDP to get full access to networking and performance as well as DOM etc? Stefan Li wired this up with devtools/mcp! Love it!

🚨 New best open-source model for agentic coding? Devstral (by Mistral AI & @allhands_ai) just hit 46.8% on SWE-Bench Verified, topping all other open weights models. - Low hardware requirements - Open weights - Test-loop agent focus Bap breaks down the news for us ➡️

Running on consumer hardware is especially exciting because it pushes forward the accessibility and democratization of powerful models.

I recently sat down with Anton Arhipov from JetBrains to talk about IntelliJ IDEA’s shift toward becoming an AI-powered IDE. He walked me through some of the intelligent plugins, like context-aware code suggestions and machine learning–driven refactoring, that are fundamentally

Traceability is becoming more important as AI systems scale, and graphs definitely make data flows clearer.

Thought Codex got better.. 🫠