brandonpurcell

[Max Markham @MaxMarkhamNFL]

This clip just got an instant induction into the Chicago Hall of Fame.

The fact that it looks like a 90’s home video makes it so much better.

[Min Choi @minchoi]

OpenClaw changed the agentic AI game.

People are building insane use cases, major companies are building on it.

There's a major shift.

10 examples👇Bookmark this

[CG @cgtwts]

Claude users right now:

[Justine Moore @venturetwins]

Incredible clip on how @karpathy uses OpenClaw to run his house via texts.

You can ask agents to find connected hardware at your home (like Sonos speaker), and they'll search the network + hack in for you 🤯

You can control music, lights, HVAC, security...w/o writing any code.

[Aakash Gupta @aakashgupta]

There are 6 levels of making Claude Code run autonomously, and most people are stuck on Level 1.

Level 1: Kill the permission prompts. Run claude --dangerously-skip-permissions. One flag. Now it stops asking “Can I edit this file?” every 30 seconds while you’re checking Slack.

Level 2: Context window management. Claude Code now supports 1M tokens. Use /clear between tasks. Run /compact at 60% usage instead of waiting for auto-compaction to fire at 90% when the model is already forgetting your instructions.

Level 3: Subagents. The reason it stops at 15 minutes: everything runs in one context window. Subagents run in separate contexts. Build a looping todo command, each task executes in its own window. Builds, tests, and git operations never touch the main conversation. 2+ hours autonomous with zero intervention.

Level 4: Ralph Wiggum loop. Official Anthropic plugin. Claude works, tries to exit, a Stop hook blocks the exit, re-feeds the same prompt. Each iteration sees modified files and git history from previous runs. One developer ran 27 hours straight, 84 tasks completed. Geoffrey Huntley ran one for three months and built a programming language with a working LLVM compiler.

Level 5: Karpathy’s AutoResearch. On March 7, Karpathy pushed a 630-line script to GitHub and went to sleep. Woke up to 100+ ML experiments completed overnight. 25K stars in five days. The difference from Ralph: structured eval loops. Define a metric, run, measure, analyze failures, improve, repeat. One Claude Code port took model accuracy from 0.44 to 0.78 R² across 22 autonomous experiments.

Level 6: VPS + OpenClaw for 24/7. Your laptop lid closing kills everything. Run Claude Code on a VPS inside tmux. Detach, close your laptop, come back tomorrow to a finished diff. OpenClaw (247K GitHub stars) takes it further: a persistent gateway connecting LLMs to your real tools, running 24/7 across messaging, email, git, and calendars. Jensen Huang at GTC called it “probably the most important release of software ever.”

The unlock at every level is the same: give Claude a way to verify its own work.

[AWS Community Builders Blogs @aws_cb_blogs]

New post by @SoPolliog

I Benchmarked TimescaleDB vs ClickHouse vs MongoDB for Observability Data - The Results Surprised Me
https://t.co/SOgfaDCo8T
#aws #database #opensource #performance

[Mike Freedman @michaelfreedman]

Introducing TigerFS - a filesystem backed by PostgreSQL, and a filesystem interface to PostgreSQL.

Idea is simple: Agents don't need fancy APIs or SDKs, they love the file system. ls, cat, find, grep. Pipelined UNIX tools. So let’s make files transactional and concurrent by backing them with a real database.

There are two ways to use it:

File-first: Write markdown, organize into directories. Writes are atomic, everything is auto-versioned. Any tool that works with files -- Claude Code, Cursor, grep, emacs -- just works. Multi-agent task coordination is just mv'ing files between todo/doing/done directories.

Data-first: Mount any Postgres database and explore it with Unix tools. For large databases, chain filters into paths that push down to SQL: .by/customer_id/123/.order/created_at/.last/10/.export/json. Bulk import/export, no SQL needed, and ships with Claude Code skills.

Every file is a real PostgreSQL row. Multiple agents and humans read and write concurrently with full ACID guarantees. The filesystem /is/ the API.

Mounts via FUSE on Linux and NFS on macOS, no extra dependencies. Point it at an existing Postgres database, or spin up a free one on Tiger Cloud or Ghost.

I built this mostly for agent workflows, but curious what else people would use it for. It's early but the core is solid. Feedback welcome.

https://t.co/IPhieopOSP

[Peter Yang @petergyang]

"I woke up and he already built a website, created a product, set up Stripe, and launched."

Here's my new episode with @nateliason on how he set up his OpenClaw bot, @FelixCraftAI, to build a business that made $14,718 in 3 weeks.

We talked about:

✅ The 3-layer memory system that prevents your bot from forgetting things

✅ How to run 5 chats with your bot at the same time

✅ How to safely give your bot Stripe, Vercel, and X access

Some quotes from Nat:

"Every time Felix asks me to do something, I ask: Can I remove this bottleneck so you never have to ask me this again?"

"I'm never in Claude Code or Codex myself anymore. I just tell Felix to do it."

"If you get the memory system and the proactivity right, it solves 90% of the frustrations that most people run into."

📌 Watch now: https://t.co/Y01F5eC7yW

Thanks to our sponsors:

@linear: The AI agent platform for modern teams https://t.co/lI40xrrDsr

@Replit: From 0 to full stack app in 2 min. https://t.co/w6kab0zMqN

[Jesse Warden @jesterxl]

Gen Z programmers, breathe a sigh of relief. Millenials devs, it's all good. Your Gen X & Boomer elders have seen this "X technology will replace programmers" before. Jesse Warden has got you.

[Jeremy Fowler @JFowlerESPN]

Chicago

[Matthew Berry @MatthewBerryTMR]

This is really well done @NFL... the Bears with one of the most epic comebacks of all-time

[Dan Go @CoachDanGo]

Over the past 5 years, I've grow my audience to 3.5M, reduced my biological age, and 8x'ed revenue.

A big reason is a science based goal setting system I call the Personal Upgrade OS.

Use this framework before the year starts to turn your goals into reality:

[Payton Halas @PaytonHalas]

you have to be a generational hater to not think this video goes fucking hard

[Ricky O'Donnell @SBN_Ricky]

Into the wind. To a receiver with a half-step of separation. I don’t think I’m exaggerating to say that’s the best throw by a Bears QB of my lifetime.

[Awful Announcing @awfulannouncing]

D.J. Moore with the post-game interview of the year following the Bears' epic win over the Packers Saturday night.

[NFL @NFL]

AN ENDING YOU HAD TO SEE TO BELIEVE

@ChicagoBears

[Mike Freedman @michaelfreedman]

There's an old systems adage: every time you scale by another 1 - 2 orders of magnitude, something new breaks or becomes the bottleneck. Not necessarily because you were unaware, but just hadn't prioritized before. Premature optimization considered harmful (or at least wasteful) and all that.

We're now spinning up databases so frequently on @TigerDatabase Cloud that our orchestration deployer started to become that bottleneck. The good news is that careful systems performance engineering still goes a long way.

Here's the before/after in orchestrator memory consumption after the latest deploy. 📉

[Branko @brankopetric00]

We chose PostgreSQL over MongoDB for our analytics platform.

The context:
- 50GB of time-series data daily
- Complex queries with joins across multiple dimensions
- Team had more SQL experience than NoSQL

MongoDB seemed obvious for scale, but:
- Query complexity made aggregation pipelines unwieldy
- Horizontal scaling wasn't needed yet
- PostgreSQL's JSON support gave us flexibility
- TimescaleDB extension handled time-series perfectly

18 months later: PostgreSQL handles 2TB with sub-second queries. Sometimes boring technology wins.

[Chubby♨️ @kimmonismus]

The fastest-adopted technology in human history – AI.

800 million weekly active users in under 3 years.