DAP iQ

@GohilHardy @grok give us the expert distribution playbook for X, reference the source code that powers X, make no mistakes 🙏

The simple distribution routine: 1. Build something small 2. Post about it 3. Reply to everyone 4. Repeat the next day 5. Join other conversations 6. Share lessons learned 7. Stay visible daily

Started adding SQLite to pretty much every project im working on lately. It helps with memory management, stateful integration testing, and lives all on the local filesystem.

The latest structure thats been working well for codebases with ~30k+ LOC: - .github > custom agents, instructions, prompts, skils - Backlog > json backlog with epics, tasks, checkpoints - Memory > project pinned md memories - Specs > short-lived feature descriptions for backlog - System > machine-readable project contracts Then using a command workflow for Init > Plan > Implement operations.

Just stumbled into @github copilot's cross-workspace internal memory store, ended up extracting and standardizing agentic coding patterns from 36 of our .NET projects. Huge timesaver!

@forwarddeploy Building ironexec ai right now, local LLM operator runtime with macOS integration

Building your own version of OpenClaw or productivity tool that uses agents? Want free xAI API credits to supercharge it with Grok? Reply below (or DM if stealth mode) Hackathon MVPs, side projects, wild experiments - let’s see ’em all! 🦞

@n0w00j There another one I’ve seen called Plaud

this could’ve been an app

@adamdotdev It’s best to focus on projects that you’re passionate about. I like finding areas where AI is still fuzzy without context/logic, then start experimenting with it. Agents are slowly replacing apps, you can embed all your domain expertise into the system to make it more rewarding.

Programming was deeply satisfying work to me. Work for hours/days before getting the payoff of the code working well on your machine. I’m feeling so much friction now to open the editor and do this kind of task by hand, but also increasingly depressed with the nature of work in an AI assisted dev workflow. Back and forth prompting seems to eat at my soul. Need to find a balance that brings back some of the toil.

How it started: setup an AI assistant on my mac mini using local @ollama LLMs and @openclaw. 7 days later we built a multi-model LLM inference lab. - 475 sessions / 16k messages - 22 model tuning experiments - 19 instrumented tools Full cron replay, conversation tracking, token cost, telemetry, and cross-provider model benchmarking.

@asaio87 Do consulting, forget the course, sell the knowledge to orgs willing to pay

I finally got the ways to make money with OpenClaw: - sell a course about it - sell a membership in a private group talking about it - sell a service for setting up OpenClaw in 30 seconds these are the only way i have seen where people actually make money with it dont you find it weird ?

@thejustinwelsh Felt this way ten years ago

Ten years from now, having a solo business will feel as normal as having a corporate job.

@MatthewBerman Openclaw.json will show you. Subagents were defaulting to my primary model until I started overriding it in the cli or try to set through config. Also this is a great time for you to build an observability dashboard so you can evaluate/debug without guessing.

It's frustrating how difficult it is to tell which model is being used by OpenClaw. It will frequently hallucinate and even /status will be wrong.

@r0ck3t23 1000% - grid can’t handle this amount of energy, big tech has tons of electricians migrating from datacenter to datacenter building them out, electricity is the bottleneck

Elon Musk just identified the next crisis in AI. It’s not a shortage. It’s an unusable surplus. Musk: “By the end of this year, chip production will outpace the ability to turn chips on.” For three years the world was starved for silicon. Every lab, every government, every company racing to secure the chips that determine who wins the AI era. That bottleneck is ending. A new one is replacing it. Musk: “The chips are going to be piling up and not be able to be turned on.” Billions of dollars of the most advanced AI hardware ever built. Sitting dark. Not because the chips don’t work. Because there isn’t enough electricity to run them. You can’t print a power plant the way you print a chip. The fabrication plants scaled. The grid didn’t. And now the most valuable hardware in history is about to hit a wall that no amount of capital can instantly solve. Compute is about to become abundant. Electricity is about to become the most valuable commodity on earth. Three years obsessing over silicon yields. Physics doesn’t care about your chip architecture if your data center can’t pull enough megawatts. The war isn’t about who can manufacture the most silicon anymore. It’s about who has the raw power to plug it in. Whoever solves energy first doesn’t just win. They own the infrastructure everyone else needs to compete. The losers stack useless chips in warehouses waiting for power that never arrives. We built a trillion dollar engine and forgot the fuel. That’s the AI race right now.

@hasantoxr I mean at this point aren’t we all running qwen3 locally for embeddings, reasoning, code - its in a class of its own in comparison to other local models, and they push updates so quickly

🚨 Alibaba just quietly dropped a vector database that destroys Pinecone, Chroma, and Weaviate. It's called Zvec and it runs directly inside your application no server, no config, no infrastructure costs. No Docker. No cloud bills. No DevOps nightmare. Built on Proxima, Alibaba's battle-tested vector search engine powering their own production systems at scale. The numbers don't lie: → Searches billions of vectors in milliseconds → pip install zvec and you're searching in under 60 seconds → Dense + sparse vectors + hybrid search in a single call And it runs everywhere: → Notebooks → Servers → Edge devices → CLI tools 100% Opensource. Apache 2.0 license. This is the vector DB the RAG community has been waiting for production-grade performance without the production-grade headache. Link in the first comment 👇

@linuz90 @openclaw I built a UI just to streamline agent/session log extraction from OpenClaw and ollama, I was able to dig into the exact timeline of events in order to refine my processes. Worth it from the learnings alone.

I really don’t get all these @openclaw “mission control” and dashboard projects. It defeats the purpose. I now have this brilliant assistant with amazing context that does all sorts of things, reports back, proactively pings etc. The fact that it doesn’t have a UI *is a feature* for me. I can relate to the nerd urge to build a dashboard for it, but I feel like not having one is kinda the point?

@RoboCanvas @TeksEdge What are we at 2-4 tokens per text character? Not a bad deal for simple tasks, but minimum of 30-60k context would be better for longer flows.

@TeksEdge 16K context is not enough. 256K kinda works. 1M or more is where its at.

🚀 I just hosted MiniMax-M2.5 UNCOMPRESSED (full BF16, 230B total / 10B active MoE, 457 GB weights) on a single Mac Ultra 512 GB! Fits with ~50 GB left for OS + modest KV cache (perfect for 4k-16k ctx). Using native MLX — zero swapping, max quality local inference. 16K context window @ 30 tps. Frontier model at home, no API. 📊 Latest AI Coding Benchmarks 🏆 SWE-Bench Verified • MiniMax M2.5: 80.2 • MiniMax M2.1: 74 • Claude Opus 4.5: 80.9 • Claude Opus 4.6: 80.8 • Gemini 3 Pro: 78 • GPT-5.2: 80 ⚙️ SWE-Bench Pro • MiniMax M2.5: 55.4 • MiniMax M2.1: 49.7 • Claude Opus 4.5: 56.9 • Claude Opus 4.6: 55.4 • Gemini 3 Pro: 54.1 • GPT-5.2: 55.6 🖥️ Terminal Bench 2 • MiniMax M2.5: 51.7 • MiniMax M2.1: 47.9 • Claude Opus 4.5: 53.4 • Claude Opus 4.6: 55.1 • Gemini 3 Pro: 54 • GPT-5.2: 54 🌍 Multi-SWE-Bench • MiniMax M2.5: 51.3 • MiniMax M2.1: 47.2 • Claude Opus 4.5: 50 • Claude Opus 4.6: 50.3 • Gemini 3 Pro: 42.7 🌐 SWE-Bench Multilingual • MiniMax M2.5: 74.1 • MiniMax M2.1: 71.9 • Claude Opus 4.5: 77.5 • Claude Opus 4.6: 77.8 • Gemini 3 Pro: 65 • GPT-5.2: 72 ✨ VIBE-Pro (AVG) • MiniMax M2.5: 54.2 • MiniMax M2.1: 42.4 • Claude Opus 4.5: 55.2 • Claude Opus 4.6: 55.6 • Gemini 3 Pro: 36.9 #MiniMaxM25 #LocalLLM #AppleSilicon

@alexhuth @TeksEdge You might be disappointed with it if you don’t have clear goals you’re looking to accomplish. Unless you have the time to fine tune a local LLM with a explicit structured workflow.

@TeksEdge Could you post a video of what that looks like in real world. I am seriously considering getting a Mac Studio but 16k context seems tiny compared to what I usually use.

@ashpreetbedi Well done, this is exactly what we’ve found.

@nummanali @grok does Codex have any restrictions against OpenClaw in its TOS?

Explicitly confirmed, no authorised usage of Claude subscription in: - OpenClaw - Pi Agent - OpenCode - Any 3rd party tool - Agents SDK No OAuth flow is allowed bar within Claude official tools If you do - you’re at high risk of a ban Was good while it lasted, be careful!

@robzolkos @grok does this impact OpenClaw

Major Claude Code policy clear up from Anthropic: "Using OAuth tokens obtained through Claude Free, Pro, or Max accounts in any other product, tool, or service — including the Agent SDK — is not permitted"

@johann_sath Spawning sub-agents for working in parallel behind the orchestrator has been a useful pattern, helps conserve space in the primary since it’s doesn’t need to be the one doing tool calls. Keep your models tuned and instructions explicit, or you’ll encounter drift.

i told my OpenClaw to stop doing work "you are the orchestrator. your time is too valuable to execute tasks. anytime i ask you to do something, spin up a subagent for it. me and you just plan. the army of subagents builds." now when i say "build this website" it: 1. spins up a worker in a docker container 2. writes the code 3. runs the build 4. pushes to github 5. deploys to production i never see a terminal. i just get "done ✅" my AI doesn't do tasks anymore. it runs a team that does.