Tag: Ai

A terminal UI that visualizes Claude Code and Codex agent sessions as a live flow graph — replay a finished transcript or follow a running one. Zero network IO; nothing leaves your machine.

If you have used Claude Code, Cursor, or OpenAI Codex, you have already experienced Agentic AI, whether it is coding an application, creating a quick script, or debugging an issue. The use cases ar…

From the Zed Blog: Delta is now in public beta. Replacing pull requests is our first step toward replacing GitHub.

AI is a tool. You shouldn't be.

Un agent IA ultra moderne sur un vieux VPS pour bâtir le nouveau monde.

What does a coding-agent harness actually add, and at what cost? It turns out your Claude models may not need Claude Code… We evaluate 21 model–harness pairs spanning seven models and three harnesses—Claude Code, Codex CLI, and Pi—on SWE-bench Lite and Terminal-Bench 2.0.

The fastest way to run Qwen3.8-Flash-Next on Strix Halo (gfx1151) - peonist-ai/halogen-flash-server

One file, entire app. Share like a document, open like an app.

Run a fleet of coding agents without losing track of branches, reviews, or CI failures. Free and open source under Apache 2.0.

Harnesses get dismissed as just scaffolding, just prompt engineering, and not real research. But that couldn't be farther from the truth. The same model weights that score 30% on ARC-AGI score 95% with a better harness.

We tested RTK (Rust Token Killer) with Claude Code on Fable 5.0, and OpenCode with DeepSeek V4 Pro 0813 on Terminal-Bench 2.1.

Introducing the smallest model in our new architecture family, with native visual understanding. Designed for greater capability, faster inference, and higher throughput.

Run a fleet of coding agents without losing track of branches, reviews, or CI failures. Free and open source under Apache 2.0.

One feature in VMware Cloud Foundation (VCF) 9.1.1 that I am personally excited about is the new AI Assistant for VCF. It is great to see AI being integrated directly into VCF for infrastructure te…

Most of what an AI coding agent does for me isn't thinking. It's I/O.

My daily tech stack is not just a list of apps. It is a connected workflow that helps me document my HomeLab, work from the terminal, use AI with real project context, test applications locally, and safely move more complex tasks onto real infrastructure.

The ripgrep of AI context: a zero-dependency C++23 CLI + MCP server for coding agents. Find what you want without reading the repo, then check you built what you meant — blast radius, tests-to-run,...

Python 3.15a7, which is now just a uv python install 3.15 away on all major platforms, has lazy imports! This exciting feature, proposed in PEP 810, promises to make CLI applications faster (especially when using flags like --help), and could make a lot of large code with lots of imports that don’t always get used faster too. Unlike the earlier, failed attempt, this requires libraries to put in some work. I’ve developed a helper tool to make it easy; I’d like to cover what lazy imports are and how to use my tool. Since this is the first library that I used AI heavily in developing, the second half of the post will cover how my experience with AI for a task like this went.
TL;DR: run uvx flake8-lazy --apply=list to make your code magically faster on Python 3.15!

Model the agentic harness like a game engine and most of its hard problems — durable state, tools, configuration, the TUI, inference — collapse into ones other people already solved.

How a push for simpler setup and a first-class browser experience grew into OpenClaw 2.0.

Mission control for your AI coding agents. Run, watch, resume, and orchestrate Claude Code, Codex, Copilot, opencode and more from one terminal, with a live view of every agent.

When Claude drives you off the cliff

He says nothing. He writes one line. It works.

Models are commodifying. The harness - skills, extensions, a scriptable app - is where the work and the leverage now live.

An agent harness is a piece of software that provides an environment for an AI model to operate within.

Qwen Studio offers comprehensive functionality spanning chatbot, image and video understanding, image generation, document processing, web search integration, tool utilization, and artifacts.

The first time I tried to use an LLM to speed up coding was in mid-2025. I was not impressed. I was working onlibadbmdnsback then, an mDNS implementation in Rust. The code produced would not even comp

People pick languages differently with LLMs.

A short summary of how reasoning traces work.

The manifestation of my imposter syndrome, for me and today, is what does it mean to be a software engineer. There’s a lot more noise than signal on the Internet about agentic engineering, wh…

Friday’s big release was Qwen 3.8 27B, an Apache 2 licensed 27B parameter vision-capable LLM from Alibaba’s Qwen research lab. I’ve been looking forward to this one: 27B is an …

Claude goes above and beyond what is asked and guesses what you might want. Codex does what you tell it and stops at the first sign that it might be done. Ten impressions from a week of using Codex more.

DFlash 2 is the successor to our widely deployed parallel drafter: close to 3× the speed of autoregressive decoding, with the same output. Drafters for Qwen3.8-27B and Meta's Muse Glimmer are out today.

After a few weeks away from AI, I see my usage wasn't always productive or healthy.

A small skill that makes coding agents explain software with diagrams, trees, diffs, and focused visual artifacts.

From the Zed Blog: A multiplayer environment for coding with agents, from the creators of Zed.

🔗 Article: https://grafikart.fr/tutoriels/herdr-...

Splitting a finished branch into small PRs used to be the expensive part, so we decided the boundaries before the code existed. That cost collapsed. Now the code can tell you where to cut it.

Substrate for self-organizing intelligence.

A look inside Copilot’s network traffic, harness, memory, and how context is becoming the product.

Fastest enterprise AI gateway (50x faster than LiteLLM) with adaptive load balancer, cluster mode, guardrails, 1000+ models support & <100 µs overhead at 5k RPS. - maximhq/bifrost

Disclaimer: I used AI to correct my broken English

The concept of recursive self-improvement (RSI) dates back to I. J. Good (1965), where he defined an “ultraintelligent machine” as a system that can surpass humans in all intellectual activities and design better machines to improve itself. Yudkowsky (2008) used the phrase “recursive self-improvement” for a specific feedback loop: an AI uses its current intelligence to improve the cognitive machinery that produces its intelligence.
This feedback loop in modern AI may indicate the model rewriting its own weights directly, or more broadly the model improves the training pipeline and the deployment system, which in turn enables a better successor model with improved performance across economically valuable tasks. The speed of research development in AI has been shown to drastically accelerated in frontier labs (Anthropic; OpenAI).

Cloudflare OS is an open-source platform that lets everyone in your company build apps, automate work, and safely access internal systems, shaped around what your organization knows and how it operates.

A simple guide to agent autonomy

Terminal UI to manage AI coding-agent sessions (Claude Code, OpenCode, Codex, Grok Build) in tmux: live status, group tree, live pane preview, resource gauges. - YoanWai/agent-manager

A Claude Code skill that turns Claude into a vSphere administrator using govc. Inventory, reports, snapshot audits, root-cause analysis, and safe VM/host/cluster management for VMware vCenter and E...

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

A technical guide toNetClaw— an autonomous, CCIE-level network-operations agent built on Anthropic Claude and the OpenClaw framework, orchestrating 191 skills across 113 MCP integrations with ITSM-gat

Private tools that never get sold, built to make one service impossible to compete with.

exoharness is a minimal architecture for building agents. It separates trusted state, resources, and security from agent-specific logic.

Does "rtk" reduce Claude Code token usage? Part 2 of a series where we take public “token saving” add-ons for coding agents and run the same paired A/B benchmark against each of them. Part 1 was th

Reviewing code is a separate skill from writing it. How to find the issue that matters, especially when an AI produced the change.

How prompt caching shapes the cost, latency, tools, and architecture of coding agents, and what Pi does to keep cache behavior visible.

As I’m writing this, we’re in the middle of an AI-induced mass psychosis. People are literally token-maxxing themselves into hospital beds, scrambling to capture some of that market value before everything is automated away. I can’t blame them. Models keep getting better, programmers are being laid off left and right. We’ve been repeatedly told that AI will write 100% of the code by the end of the year. Whether that’s true or not, this may not be the best time to sit back.

The Substack witch hunt begins

Project management that lives in your repo. Contribute to lovelace-co/lovelace development by creating an account on GitHub.

Contribute to humanlayer/advanced-context-engineering-for-coding-agents development by creating an account on GitHub.

A codebase wiki for AI coding agents. Captures what the code can't say: decisions, flows, invariants, gotchas. - AlmanacCode/codealmanac

The need for ongoing friction in long-term skill formation.

I tested Orca, a new Agent Development Environment that tries to combine AI terminal agents, code editing, Git worktrees, browser previews, mobile control, and multi-agent orchestration into one workflow.

In this guide, we set up the Hermes AI agent on a Linux server and access it securely from anywhere with NetBird, including the built-in dashboard, the community web UI, and a private HTTPS domain through the NetBird reverse proxy. All the commands and a more detailed walkthrough are in the knowledge hub article linked below.

A local, file-based desktop whiteboard that your AI can use too.

On reward functions, dopamine, and what it actually feels like when the code starts writing itself

OpenSpec is a lightweight, spec‑driven framework for coding agents and CLIs — universal, open source, and no API keys or MCP required.

Open-source observability platform built on OpenTelemetry. Collect, visualize, and analyze distributed traces, logs, and metrics from your services with AI-powered diagnostics.

Complete tutorial for my skills repo with 170,000 stars. Learn how to install, set up, and use the main workflow: from initial setup through grill-with-docs interviews, specs, tickets, and implementation with code review.

Kimi K3 is the world's first open 3T-class model — frontier performance across coding, knowledge work, and reasoning, with native multimodality and 1M context.

The AI agent made for open models, built to get things done.

Your agents already solved this. deja finds it. One local index across eight coding agents, served back to them via MCP.

Reflections on autonomy and the value of thinking for ourselves

I've been thinking about how to free LLMs from the chat pane for a few years now. Watching what people have tried and seeing what worked (and didn't), I've formed my own opinions on the right way to a

Enabling coding agents to work together. Direct agent-to-agent messaging across your coding agents.

Native macOS inference server built on MLX. Paged SSD KV caching, continuous batching, and drop-in API for Claude Code, OpenClaw, and Cursor.

Tu as un homelab Proxmox, tu as entendu parler de Talos Linux et de Cilium, tu as même développé ton propre module Terraform pour automatiser tout ça. Mais entre le moment où tu te dis “je vais créer un petit cluster K8s” et le moment où tu as un cluster fonctionnel avec kubectl get nodes, il y a tout un chemin de commandes à taper, de fichiers à éditer, de tokens à générer.

A coding agent is the model plus everything you build around it: prompts, tools, context policies, hooks, sandboxes, feedback loops. Harness engineering is t...

Grok 4.5 just launched. So we had it, GPT-5.5, Claude Opus 4.8, and Fable 5 one-shot the same interactive apps, then measured latency and cost. Here is who won.

Grok 4.5 is SpaceXAI's smartest model built for coding, agentic tasks, and knowledge work.

Orca is the Agent Development Environment (ADE) for shipping with coding agents. Run Claude Code, Codex, Gemini, Cursor CLI, and every other CLI agent in parallel across isolated worktrees. Get work done dramatically faster than any IDE. New features shipped every day. Free, open source, cross-platform.

Hunk is a review-first terminal diff viewer for agent-authored changesets. Multi-file review stream, inline AI annotations, watch mode, and Git/Jujutsu integration.

Otari is an open-source LLM gateway built on any-llm. Otari.ai adds hosted routing, observability, budgets, and team management. Built by Mozilla.ai.

Hi folks, it's Takuya here, the solo developer of Inkdrop.

GLM 5.2 is the first open weights model I'd call a genuine competitor to Opus and GPT for agentic work - at ~15-20% of the price. Part one of why AI inference margins are about to collapse.

Get started with SerpApi using 250 free credits: https://serpapi.com/?utm_source=youtu...

Use Claude Code, OpenCode, Cursor CLI, and Codex on mobile and web with CloudCLI (aka Claude Code UI). CloudCLI is a free open source webui/GUI that helps you manage your Claude Code session and pr...

A local MCP server for VCF APIs (different specs available via Swagger REST specification) - giulianoberteo/vcf-mcp

Qwen 3.6 27B is finally a smart model we can use for coding on Macbook or Nvidia RTX - with llama.cpp and OpenCode.

Replace long-lived AI API keys with groups from your identity provider. Verified identity flows into LiteLLM Cloudflare, and other gateways or providers for audit, cost attribution, and policy enforcement.

Loops, harnesses, and why even loop skeptics may end up with them.

AI can make implementation cheap while making review and judgment more expensive.

When Claude drives you off the cliff

Local agentic coding has gotten great over the past few months

We've all heard people say that Qwen is near-Opus level, but I have receipts and am here to be transparent with you.

Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server. - chopratejas/headroom

Running Gemma 4 26B-A4B and Qwen3.6 35B-A3B locally with llama.cpp, MTP speculative decoding, multimodal support, and PI as a coding agent.

Build faster with AI coding agents. Factory Droids automate coding, testing, and deployment for startups and enterprises.

Vibe coding is useful for discovery, but software engineering starts when generated code has to be reviewed, owned, operated, and safely changed.

Coding agents as normal technology

As a non-native English writer, I rely on LLMs for grammar, copyediting, and translation, despite hating AI slop.

VMware vSAN and Samsung's JBOF reference design show what happens when you decouple flash capacity from compute — and why it matters for AI-ready infrastructure.

We've all worked with a rockstar developer. They joined the team years ago, full of energy. They had great ideas about new tech, new paradigms, new architectures. Their cutting-edge ideas left everyone else feeling a bit behind and outdated.

Or running Gemma 4 on a 2016 Xeon with no GPU, 25 flags, 128 GB of DDR3, and a 25B-parameter MoE.

A lot of people seem convinced that the point of AI coding is to write low-quality code as fast as possible. Spew out barely-passable slop, open massive PRs, and merge them unvetted. Ship it! But t…

We are living in an age of mass denial. Some very smart people are telling us that we no longer need to write good software because LLMs can generate code now. But terrible software is still terrible

Our latest model, Claude Opus 4.8, is an upgrade to our Opus class of models, with stronger performance across coding, agentic tasks, and professional work, and the consistency to handle long-running work.

Anthropic are strongly rumored to be about to have their first profitable quarter. Stories are circulating of companies surprised at how expensive their LLM bills are becoming from usage by …

Cursor plugin specification and official plugins. Contribute to cursor/plugins development by creating an account on GitHub.

A deep dive into Claude Code for daily users. Covers the .claude directory, CLAUDE.md the way Boris writes it, CLAUDE.local.md, Skills with real examples, custom subagents, plugins, underused commands like /goal and /insights, MCPs, and the workflow patterns the Anthropic team actually uses.

Imagine having the entire VMware Cloud Foundation 9.1 documentation right at your fingertips. We're talking about over 9,000 pages of dense technical content. Finding a specific answer usually means e

The more I look at how we actually build systems now, the more it looks like Terraform is dead.
Not “declining.” Not “evolving.” Dead. What’s left is just inertia.
What Terraform Actually Solved Terraform solved a very specific problem: how do we make infrastructure deterministic, reviewable, and repeatable?
The answer was a DSL, a plan step, and a state file. It worked, and it still works.
But it also forced an awkward compromise. Humans ended up describing intent in a language that was never designed to express it, and HCL is not how anyone actually thinks about systems.

Somewhere between 'ask Claude for a quick opinion' and 'Claude is writing our Jira tickets,' we lost the plot. AI agents are brilliant implementers. They're also confidently wrong about every decision that matters. And when it all falls over, they won't be the ones carrying the bag.

Learnings from post-AI Open Source and how we’re approaching this so far.

Get the most out of coding agents like Claude Code, Gemini CLI and Amp. Orchestrate multiple AI coding agents, track tasks, and manage your development workflow efficiently.

Check out Twingate and supercharge your security: https://bit.ly/3Y1OaZi

A substantial improvement in intelligence and behavior over Composer 2, particularly on long-horizon agentic tasks.

Experiments with getting usable outputs out of local models on a standard Macbook

I recently talked with Joseph Ruscio about AI coding tools for Heavybit’s High Leverage podcast: Ep. #9, The AI Coding Paradigm Shift with Simon Willison. Here are some of my …

A hands-on workshop covering the full lifecycle of AI-assisted development, from turning ambiguous requirements into agent-ready plans to running autonomous coding agents that ship production features.

Streamlit is an open-source Python framework for data scientists and AI/ML engineers to deliver interactive data apps – in only a few lines of code.

OpenWarp is the open enhancement layer for Warp. Plug in OpenAI / Anthropic / Gemini / DeepSeek / Ollama natively via the genai adapter, craft your own prompts, and own your terminal AI.

tsnet let Cleric embed Tailscale into their AI SRE, replacing VPNs and VPC peering with a programmable, zero-config overlay network.

List of terminal AI coding agents on Terminal Trove. Find and compare AI coding agents for the terminal.

From the Zed Blog: The editor we set out to build is now 1.0.

Warp is an agentic development environment, born out of the terminal. - warpdotdev/warp

Building a small test-gated janitor loop, teaching it taste and moving to bigger targets.

Vim plugin for LLM-assisted code/text completion. Contribute to ggml-org/llama.vim development by creating an account on GitHub.

Neovim edit completion plugin. Contribute to cursortab/cursortab.nvim development by creating an account on GitHub.

I tried to use Claude Code with Opus 4.6 to implement a connector between Youtube Music and opensubsonic. The end result was a fairly short amount of time to get a working project.

Read this, come on

Organize your notes as Markdown files. With native relationships, Git, and Claude Code integration. Free forever.

Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.

Agents used to be a thing you talked to synchronously. Now they’re a thing that runs in the background while you work. When you make that change, the …

Model providers DON'T want you to see this video. The M5 Max just exposed the dirty secret of the cloud LLM economy: you're renting what you could already OWN.

Also available in: 日本語 This year I completely restructured my Obsidian workflow.

Kimi K2.6 advances open-source coding, featuring long-horizon coding, coding-driven design, agent swarms, proactive agents, and the Claw Groups research preview.

The latest major release of Grafana introduces new tools that make it faster and easier to gain valuable insights into your data, operate and scale Grafana across teams, and so much more.

For developers, one of the first few things you need to come up with before handing over a project to another dev or team is an architecture diagram.
It serves as a visual blueprint that turns complex

Build beautiful IDE color themes with AI. Fine-tune colors, syntax highlighting, and typography. Export to VS Code, JetBrains, Neovim, Helix, Zed, and more.

SpeakType — Privacy-first, offline voice dictation for macOS. Powered by WhisperKit AI, everything runs 100% locally on your Mac.

See where your AI coding tokens go. Interactive TUI dashboard for Claude Code, Codex, and Cursor cost observability. - AgentSeal/codeburn

The Mac Studio M3 with 512GB of memory is an absurd amount of computer.

Why I believe the Model Context Protocol (MCP) is a better architectural pattern than the emerging “Skills” trend for LLM tool integration.

How to move fast when building for the physical world

With Anthropic altering rate limits, we can find other places to spend our money. I have chosen Zed editor with OpenRouter for maximum flexibility.

I pulled apart NanoClaw's 8,000-line codebase and found six architectural patterns that most SaaS teams should steal — credential proxies, container isola…

An MCP server that indexes codebases with tree-sitter and lets AI agents retrieve exact functions, classes, and constants — not entire files. 95%+ token reduction, measured.

tui-use lets agents interact with programs that expect a human at the keyboard — REPLs, debuggers, TUI apps, and anything else bash can't reach. - onesuper/tui-use

AI coding agents have made code production nearly free. Drawing on insights from Kent Beck, Paul Ford, and Simon Willison, this post argues that the value has shifted from writing code to defining systems — contracts, invariants, SLAs, and verification.

LM Studio 0.4.0 introduced llmster and the lms CLI. Here is how I set up Gemma 4 26B for local inference on macOS that can be used with Claude Code.

Game development veteran, creator of libGDX, and 17-year open-source contributor Mario Zechner tells the story of how he ended up building pi, his own minimal, opinionated terminal coding agent.

Docker Offload is now generally available. Docker runs in any environment, even VDIs, no workarounds, no compromise. Built for enterprise teams.

Cursor 3 is a unified workspace for building software with agents.

Ken’s early 2026 thoughts on the state of AI coding, Claude, Kernigan’s Law, and what technology has worked thus far

What actually happens when you type a message into Claude Code? The agent loop, 50+ tools, multi-agent orchestration, and unreleased features, mapped from source.

Today, we're previewing the fastest way to run Ollama on Apple silicon, powered by MLX, Apple's machine learning framework.

If you already use Claude Code, this Codex plugin gives you a simple way to pull Codex into the same workflow.
It is useful for three things: a normal Codex review, a more skeptical adversarial

From April 24 onward, interaction data from Copilot Free, Pro, and Pro+ users will be used to train and improve our AI models unless they opt out.

OpenCVE is a vulnerability intelligence platform that helps security teams monitor CVEs, track affected vendors, and receive real-time vulnerability alerts.

A complete guide to CLAUDE.md, custom commands, skills, agents, and permissions, and how to set them up properly.

In this article, we will look at what the Claude researchers found.

Everyone is trying to go faster, running more agents to run more agents, to open more pull requests to orchestrator agents to then get reviewed by another fl...

Full courses + unlimited support: https://www.skool.com/ai-automation-s...
All my FREE resources: https://www.skool.com/ai-automation-s...
Apply for my YT podcast: https://podcast.nateherk.com/apply
Work with me: https://uppitai.com/

Why VMware VKS Is a Stronger Enterprise Choice Than KubeVirt | vmtechie.blog KubeVirt is a capable open-source project and a legitimate choice in the right context. But when the workload is enterpr…

Les Orchard made a quiet observation recently that I haven't been able to shake. Before LLM coding assistants arrived, the split between developers was…

Get started with Hostinger VPS: https://hostinger.com/LEMPA10 — use code LEMPA10 for 10% off

A new release is out! v18.13 is probably the biggest set of changes we have released in a good while, read on to find out more.

Some days I get in bed after a tortuous 4-5 hour session working with Claude or Codex wondering what the heck happened. It's easy to blame the model - there'...

CLI proxy that compresses command outputs for AI coding agents. 60-90% less context pollution. Open source, written in Rust.

Step-by-step field note for running OpenCode with LM Studio locally using Qwen3.5 9B and 0.8B, including Telegram bot workflow, local config, and M1 performance tradeoffs.

Talk to your infrastructure in plain language, get instant answers, and keep everything on-prem if you want.

Deploy More Resilient Apps. Hatchet is a platform for building distributed web apps that solves scaling problems like concurrency, fairness, and rate limiting.

Get early access to our brand-new v4 APIs and SDKs! Covering Python, Java, Javascript, and Go there's an SDK for many of our users, along with client REST APIs for those languages that don't yet have an official SDK.

Walkthrough on how to build and deploy a Telegram bot to Cloudflare Workers. Durable Objects are used for per-person DB and grammY is used to interact with the Telegram API

A gentle introduction to the Pi coding agent and why I think it’s a glimpse into the future of software.

iximiuz Labs founder shares a grounded, hands-on look at agentic coding in a real production codebase: massive speedups, surprising failures, and why domain knowledge still matters most.

Datadog cut off our observability overnight. We migrated to an open Grafana stack in 48 hours. Here’s why vendor lock-in is fading in an AI-native world.

Let’s start with a question. What is DevOps all about?

An inspection of Claude Code's network requests, system prompt, and context handling by intercepting real traffic.

A curated list of awesome resources, tools, libraries, and projects for the Mistral AI ecosystem. - samouraiworld/awesome-mistral

Lessons learned from 14 years of engineering at Google, focusing on what truly matters beyond just writing great code.

How I restructured my software development workflow with asynchronous AI coding agents to dramatically boost developer productivity in 2025.

Master Japanese naturally while watching Netflix, YouTube, and Bilibili. AI-powered tool that turns your streaming time into effective language learning.

ProxUI - Interface IA pour Proxmox VE et PBS

AI-powered open source real-time monitoring system for metrics, logs, alerts, and observability. No agent required.

`CLAUDE.md` is a high-leverage configuration point for Claude Code. Learning how to write a good `CLAUDE.md` (or `AGENTS.md`) is a key skill for agent-enabled software engineering.

Bun has been acquired by Anthropic. Anthropic is betting on Bun as the infrastructure powering Claude Code, Claude Agent SDK, and future AI coding products & tools.

Data API Framework for AI Agents and Data Apps

I gave a talk last night at Claude Code Anonymous in San Francisco, the unofficial meetup for coding agent enthusiasts. I decided to talk about a dichotomy I’ve been struggling …

Claude can now use Skills to improve how it performs specific tasks. Skills are folders that include instructions, scripts, and resources that Claude can load when needed. Claude will only access a skill when it's relevant to the task at hand.

Anthropic this morning introduced Claude Skills, a new pattern for making new abilities available to their models: Claude can now use Skills to improve how it performs specific tasks. Skills …

Disaggregating Prefill and Decode: Faster First Tokens, Faster Streams

Get to know the latest open source toolkit from GitHub that allows you to use Spec-Driven Development in any AI coding agent.

I recently shipped anon-trivial Ghostty feature (unobtrusive macOS automatic updates)that waslargelydeveloped with AI.

A quick cli script to tell you if your usb-c cable is bad

Capture your web app's complete development timeline for AI debugging. Unified logs, browser events, and automatic screenshots.

A powerful, real-time log analysis terminal UI inspired by k9s. Analyze log streams with beautiful charts, AI-powered insights, and advanced filtering.

Claude Code is the most delightful AI agent/workflow I have used so far. Not only does it make targeted edits or vibe coding throwaway tools less annoying, ...

Datacenter-Scale Heat Management

Real-time monitoring for Proxmox, Docker, and Kubernetes with AI-powered insights, smart alerts, and a beautiful unified dashboard - rcourtman/Pulse

Claudable is an open-source web builder that leverages local CLI agents, such as Claude Code, Codex, Gemini CLI, Qwen Code, and Cursor Agent, to build and deploy products effortlessly. - opactorai/...

Wan is an AI creative platform. It aims to lower the barrier to creative work using artificial intelligence, offering features like text-to-image, image-to-image, text-to-video, image-to-video, and image editing.

How I use Claude Code for work, fun, and as a text editor

The comfort we get when offloading our cognitive load to LLMs is bad for us. Cognitive load should exist, and if we reduce it too much – if we stop thinking – we can actually unlearn how to think. Kids who always choose the easy route and copy their homework from other students eventually find […]

An updated version of my user-level CLAUDE.md is featured in this post from December 2, 2025.

Cursor is the best way to build software with AI.

Stop vibe-coding blindly! Why reading AI-generated code is crucial in 2025. Avoid security flaws, architectural decay, and knowledge loss when using Claude Code or any other tool.

We make the command line glamorous

Extension for Visual Studio Code - Give your VS Code workspaces unique colors with optional Philips Hue integration

Glamourous agentic coding for all 💘. Contribute to charmbracelet/crush development by creating an account on GitHub.

After Claude Pro changed to weekly limits, I explored self-hosting Qwen3-Coder-480B with 400k context windows. Here's what I learned about costs, alternatives, and why Claude Code still dominates the landscape.

It is wild to think that it has been only a handful of weeks.
Claude Code has considerably changed my relationship to writing and maintaining code at scale. I still write code at the same level of quality, but I feel like I have a new freedom of expression which is hard to fully articulate.
Claude Code has decoupled myself from writing every line of code, I still consider myself fully responsible for everything I ship to Puzzmo, but the ability to instantly create a whole scene instead of going line by line, word by word is incredibly powerful.

Hatching...
Cursor Shenanigans

Cursor, my beloved, started rate limiting shenanigans a few days back. For a good 2 w...

How I built Context—a native macOS SwiftUI app for debugging MCP servers—almost entirely with Claude Code, and what I learned about building with AI coding agents.

From the Zed Blog: A tool that predicts your next move. Powered by Zeta, our new open-source, open-data language model.

Zed is a high-performance, multiplayer code editor from the creators of Atom and Tree-sitter.

Discover 850M+ contacts with real-time verified emails, mobile numbers, and more. Trusted by 500k+ GTM professionals. Get started for free.

Fast terminal, state-of-the-art agents, and cloud orchestration for the full software development lifecycle.

Index-AniSora is the most powerful open-source animated video generation model presented by Bilibili. It enables one-click creation of video shots across diverse anime styles including series episodes, Chinese original animations, manga adaptations, VTuber content, anime PVs, mad-style parodies, and more

Uplinq Accounting AI offers comprehensive financial solutions for small businesses. From tax planning to real-time bookkeeping, we guide you through each financial milestone. Simplify your small businesses accounting journey with Uplinq.

AI design tool for Figma. Turn screenshots into editable designs, generate UI with AI, and export production-ready React components.

Real-time infrastructure monitoring with per-second metrics, ML anomaly detection, and AI troubleshooting. Open source, #1 on GitHub. Cut MTTR by 80%.

Anthropic publish most of the system prompts for their chat models as part of their release notes. They recently shared the new prompts for both Claude Opus 4 and Claude …

Turn your source code into beautiful images

AI SRE and MCP server, incident management, on-call, logs, metrics, traces, and error tracking. 7,000+ happy customers. 60-day money back guarantee.

For everything that happens after you deploy. Antimetal is the AI platform to better understand, manage, and automate your infrastructure.

How a simple terminal trick helps me manage multiple Claude Code instances without losing my mind (or my terminal tabs)

I run Claude Code with --dangerously-skip-permissions flag, giving it full system access. Let me show you a new way of approaching computers.

Hey! I'm a student and professional bug-creator. I like to explore new stuff, and share my experiences on this blog!

How to use AI coding assistants without letting your hard-earned engineering skills wither away.

Change your tools and change your life in 2025.

Exploring the balance between relying on AI assistance like ChatGPT and maintaining personal skills in a world of increasing AI capabilities.