Eric Clemmons

How GSD turns Claude into a self-steering developer The speed at which ClawdBot MoltBot OpenClaw climbed in popularity was quite phenomenal, and for good reason: It has an audience beyond the developer space, especially those who just want to experiment with digital assistance. Todays post continues my journey with GSD, but the LLM agent landscape is changing rapidly and this shooting star caught everyone’s eye. At its core, OpenClaw is a locally hosted, open-source AI assistant that runs on your computer or server. You talk to it through apps you probably already use, like WhatsApp, Telegram, Discord, Slack, Signal, or even iMessage. And it really “does stuff” — sometimes too well. It is a gateway service that maintains WebSocket connections and includes a smart orchestration layer for working with LLM agents. Stay connected for industry’s latest content – Follow Deepthi Talasila #DevSecOps #ApplicationSecurity #AgenticAI #CloudSecurity #CyberSecurity #AIinSecurity #SecureDevOps #AppSec #AIandSecurity #CloudComputing #SecurityEngineering #ZeroTrust #MLSecurity #AICompliance #SecurityAutomation #SecureCoding #linkedin #InfoSec #SecurityByDesign #AIThreatDetection #CloudNativeSecurity #ShiftLeftSecurity #SecureAI #AIinDevSecOps #SecurityOps #CyberResilience #DataSecurity #SecurityInnovation #SecurityArchitecture #TrustworthyAI #AIinCloudSecurity #NextGenSecurity https://lnkd.in/g8YPff7n

1

So which models produce HTML that has the least accessibility errors, and at what cost? Mapped it out here (bottom left is the best). This is helpful for understanding which models are most compliant for those with disabilities and do it efficiently. Qwen and Gemini 2.5 Flash lead the pack. Hat tip to Ben Ogilvie for pointing me to this: aimac.ai

55

2 Comments

https://lnkd.in/gFajdc6G Great article Philipp Schmid! Senior devs are struggling to develop native agentic systems because they’re trying to code away the ambiguity. But in agentic system design, you can’t "if/else" your way to a success. At Switchboard, we’ve spent months sweating out the tug of war between predictable logic and LLM creativity inside of automated workflows. The hardest part? Knowing where to draw the line. Too much determinism builds in brittleness. Too much non-determinism gives you a "black box" that customers and architects can’t trust. My current rule of thumb: Use deterministic logic for the hard-rails (must execute logic, tenant segregation, etc) and non-deterministic logic for decisions we would normally think of a human owning on the other side of a mouse and keyboard. We want to build towards systems that are resilient, adaptive, and given enough reins to show emergent creative problem solving under controlled risk.

10

I use Claude Code all day as a Product Manager. I built 18 custom Claude Code commands that I'm giving away for free. See comments for the link. I use these in Nimbalyst as its the best way to work with Claude Code - https://nimbalyst.com, but these custom commands will work however you use Claude Code: /plan - Feature planning /prd - Create structured PRDs /strategy - Strategy documents /competitive - Competitive analysis /research - Market research synthesis /customer-interview - Customer interview guides /customer-interview-simulate - Simulate customer interviews /feedback-analyze - Analyze customer feedback patterns /understand-feature - Examine codebase for feature context /mockup - Generate UI mockups /documentation - Create documentation /edge-cases - Identify edge cases /bug-report - Document bugs /triage-requests - Triage incoming requests /status - Development status updates /github-status - GitHub progress summaries /launch - Launch announcements /sales-enablement - Sales enablement materials

176

29 Comments