Twitter Bookmarks

Problem: Running multiple agents leads to jumping between multiple terminals and unknown wait times. Solution: Control the agents via a Kanban board. Display a red card when the agent is blocked. Result: One spot to monitor and interact. This is so 🔥

https://t.co/JrayLee7b2

https://t.co/UVlRDGJu5I

PSA for a CTO, Head of AI, VP/Dir of Engineering, CXO: This is going to be one of the most important "back to work" weeks of your career. You must get your team aligned on agentic dev ASAP. If you're feeling behind or overwhelmed, here are some good reads to get you inspired 🧵

17 rules for successful agentic coding (also known as the unix philosophy) https://t.co/BAAiYBIZt0

I'm Boris and I created Claude Code. Lots of people have asked how I use Claude Code, so I wanted to show off my setup a bit. My setup might be surprisingly vanilla! Claude Code works great out of the box, so I personally don't customize it much. There is no one correct way t...

playing with Gas Town it’s becoming obvious very quickly mental model drift is now the main problem how do you keep your head around a system that does days of work while you went for lunch

when I was inquiring about call stack architecture, the most common response I got that I had never tried out was "beads" looks like beads author steve yegge just launched an IDE called Gas Town for orchestrating agents you've officially piqued my curiosity Steve, will try.....

Must-Read, not just for iOS. Thomas is pushing the status quo.

The State of Agentic iOS Engineering in 2026 My perspective on AI-driven programming, workflows, and tooling https://t.co/a4xAAbK4vg

Skill file: --- name: ask-questions-if-underspecified description: Clarify requirements before implementing. Do not use automatically, only when invoked explicitly. --- # Ask Questions If Underspecified ## Goal Ask the minimum set of clarifying questions needed to avoid wr...

https://t.co/pE2pQ1isw8

I think I finally found the sweet spot for long-running AI coding tasks: GLM 4.7 feels like Sonnet, costs ~20× less, and runs forever without melting your budget. Pair it with a good agent setup and it quietly does real work. https://t.co/7xSaNY810t

@Shpigford Beads https://t.co/KgWygHcKNn

Here's a Christmas present for my operators: Lots of people are asking me for the prompt we used for our 2026 strategy interviewer. One improvement I'd make to this is requiring specific information for in-quarter projects documenting start date, end date, dependences (peo...

At @every, we created our 2026 strategic plan by building a chatbot that asks you questions and debates your answers. It decreased our planning cycle time by 60% and improved thoroughness and alignment by 3x, easily. The chatbot had access to our top-level strategic plan an...

Tim wrote about how he took his hands off the wheel and let Amp write 90% of our TUI framework. And he suspects that's why Amp now rips through our codebase. https://t.co/nSjSbeoKnW https://t.co/MqpvuPp0Ti

Announcing Arcway ⊹ ࣪ ˖ A real-time 3D engine where anyone can design a home. It is a simulated 3D world where buyers explore, change, and decide as light, physics, building rules, and real products move as one. https://t.co/mt9LhKQQTc

If you want to have Opus 4.5 at it's MAX in Claude Code Update ~/.claude/settings.json { "env": { "CLAUDE_CODE_MAX_OUTPUT_TOKENS": "64000", "MAX_THINKING_TOKENS": "31999" } } - double the output - ultrathink always on Ctrl+O will show thinking in verbose mode

Advent of Claude Day 13 - Prompt Stashing It's like git stash, but for your prompt. Ctrl+S saves your draft. Send something else. Draft auto-restores. No more copying to a scratchpad. No more losing your train of thought.

My GPT is annoying me. I want to change the instructions. What are some project/gpt instructions that've worked well for you?

Coding agents suck at using a browser. Playwright MCP burns through your context window before you even send your first prompt. That is why I built Dev Browser, a Claude Skill to let your agent close the loop without eating up tokens. https://t.co/wufLNZGSpg

Day 99 – Vibe Coding an App Until I Make $1,000,000 | Revenue: $1,926.96 https://t.co/6JMTNSedpW

We built a bug finder. We're finding serious, "let's fix that right now" issues in every codebase we run it on. Introducing Detail! https://t.co/BGgVFvRlBH

feature / request for a startup would love to get a demo video - possibly generated by AI - for new feature we ship would be awesome to create a PR and then get an ai generated demo video we can share internally or with our customers lmk if ur building this - super willi...

generate diagrams and visualize code with this command https://t.co/GZgZfyt8Uf https://t.co/YLHubjsHos

# The Context Layer The impact of AI on how organizations work is just beginning. Today humans use LLMs by pasting in text and asking for feedback. Even by next year, this will seem archaic and antiquated. AI is going to turbo-charge the productivity of information work. Sk...

This is a very good post on agent design https://t.co/pEbNUJsOme

> bezos starts a stealth AI lab > immediately ships continual learning just do the obvious thing let your customers backprop

@tekbog @beffjezos it’s not LoRA https://t.co/eUTgfP8uOO