gtm / ops management / sales book a call: https://nitter.gq/t.co/s8yZjBNADF
Joined May 2011
- Tweets26.7K
- Following954
- Followers1.6K
- Likes48.8K
Dylan retweeted
Buckle up, this is a technical one. 🤓
I run @claudeai Code across 3 PCs, and every one of them works from the same memory!
There's my daily driver (@Windows 11 and an RTX 4080, the one I'm sitting at now), an always-on PC in the corner that never sleeps, and a laptop for when I'm on the go. Whichever machine I sit down at, my AI has the same files, the same notes, and the same context it had five minutes ago on another one.
More details: SSH lets me hop from one machine to the next, Syncthing keeps the big files in sync live, and an hourly Git backup puts everything in a private repo in the cloud. @Tailscale ties it together, so it works from anywhere. The laptop is short on disk space, so it skips Syncthing and keeps a slim Git copy instead.
The always-on PC is the workhorse. It runs lilAgents routines around the clock: standup prep every Monday, a LinkedIn stats pull, a @HubSpot sync that keeps our CRM current. Some run daily, some weekly, some only on weekdays. Others wait for a trigger, like someone tagging it in @Discord, a file landing in a folder, or a build request on @GitHub. Discord doubles as my remote control, so I can talk to it from my phone.
Big jobs get split up. When I render a batch of videos for our podcast, #Retooled, both PCs pull from one shared list of chunks, so neither machine bogs down, and the final join happens where I can watch it. Audits work the same way: five specialist agents go through a Google Ads account in parallel, skeptical verifiers try to disprove each finding, and one last step writes the ranked report.
And not everything needs a model. Some jobs get Opus, some get Haiku, and some are just a script on a timer.
What's the one task you'd hand to a computer that never turns off?
#ClaudeCode #AIAutomation
Dylan retweeted
Calling all builders, tinkerers, operators, and vibe coders.
AGENTS is a Discord for people who actually run AI agents. Share your setup, steal someone else's, and show what you shipped. Any stack, any level.
It's invite-only for now. Drop an intro and we'll let you in. Link below.
Dylan retweeted
Here's exactly how we use AI to get our podcast #Retooled out every single week, automatically.
🎙️ We join a Google Meet for the conversation with our guest. Our host David records his mic with OBS via multiple separate tracks, and co-host Dylan records his mic locally. Fathom.ai records the entire meeting as a backup.
✅ Before we start, we confirm the guest's name pronunciation, business name, location, headshot and title. Then we start the episode and record the content.
⏱️ After roughly half an hour, we explain to the guest approximately when the episode will go live. We automatically save clips from the episode that we send to the guest to post when it goes live.
🖥️ A separate always-on PC edits and renders the podcast for us automatically:
• Leveling out the audio
• Cutting out the mistakes
• Adding an intro and an outro
• Grabbing a highlight from the episode and placing it at the beginning
📡 Once that's finished, it automatically gets distributed to Spotify, Apple Podcasts, all the other major podcast locations via RSS, YouTube, and the lilAgents website.
📅 It's then entered into the queue to be published the next available Friday, after every other queued episode.
🚀 On release day, it gets switched from scheduled to public, is published on every major platform, and goes live on our site. The full show notes and transcript get published there, too, and it all goes out on our social channels automatically.
Reply if you want more details.
Dylan retweeted
Our weekly team meeting runs itself.
Claude preps the agenda from last week's action items, checks what actually got done, pings owners, refreshes the scoreboard, and posts action items after the call linked to the exact moment in the recording.
Here's the whole loop 👇
Dylan retweeted
💬 @dhh: "the programming language of the future is English"
Even bigger shift is that ambiguity becomes a feature I think.
You describe intent, direction, and guide taste. Agents figure out the rest.
The scarce skill moves from writing precise instructions to knowing what good looks like.
Code → intent
At what point do we stop calling this programming?
Here's my conversation with @DHH. He is back for round 2! It was an epic fun 5 hour conversation about the future of programming, AI, Linux, and human civilization.
It's here on X in full and is up everywhere else (see comment).
Timestamps:
0:00 - Episode highlight
1:27 - Introduction
2:56 - Programming with AI agents
18:14 - How software will change
27:30 - AI impact on open source
37:21 - Building Omarchy Linux distro
47:05 - Vibe coding vs agentic engineering
1:00:06 - The end of manual programming
1:10:24 - Advice for programmers
1:22:30 - Surviving Internet Hate
1:31:46 - Programming setup for AI Agents
1:44:11 - Obsessing about speed
2:07:06 - Voice prompting vs typing
2:21:05 - Best AI coding models
2:37:55 - Best AI coding harnesses
2:50:57 - AI video generation and filmmaking
3:10:28 - Fatherhood
3:38:35 - Linux will win the desktop
3:49:51 - PewDiePie
3:59:24 - Future of programming
4:22:17 - Politics and immigration
4:53:54 - Longevity, over-optimization, and fear of death
5:05:38 - Eternal recurrence and future of human civization
Dylan retweeted
my personal favourite Will Angus bit
Will Angus, who appears in “Artificial” as Anthropic co-founder Dario Amodei, says Andrew Garfield “nailed” his performance as Sam Altman:
“He has this vocal fry in between lines, kind of when he’s thinking, that he really nailed. The walk, the head movements. It was spooky.”
variety.com/2026/film/news/a…
Dylan retweeted
💬 Paul Graham on the benefits of joining an accelerator like YC:
🤝 You get colleagues
Starting a startup is usually lonely. In an accelerator you’re surrounded by other founders building alongside you, even if everyone is working on their own startup.
🧠 You learn from other startups
If you hit a technical or operational issue, there’s a good chance another startup in the batch has dealt with it already, so you can ask how they solved it.
💸 You get a built-in early customer base
You can often sell your product to other startups in the batch, who are exactly the kind of users you want - early adopters who decide quickly.
Dylan retweeted
Opening an affiliate program for a client: a B2B engineering services company.
Their average client lifetime value is $150K. Partners earn 5% of lifetime revenue on every client they refer, so a single referral works out to roughly $7,500 over the life of the account.
Good fit: anyone with an audience or network of founders, CTOs, or product leaders at US companies up to 50 people, including funded startups, agencies, and scale-ups.
Full program terms and the client's identity are at the application link: docs.google.com/forms/d/e/1F…
Dylan retweeted
Here it is: how to set up your AI-native company from scratch.
Includes:
• Picking a name
• Account(s) set up
• LLC/company formation
• Selecting an AI Harness
• Hardware / PCs
• Platform / data ownership vs. renting
This is @lilAgents' playbook!
Dylan retweeted
This is Knap. It's a new language I created that turns data into Markdown. The syntax should feel familiar and comes with wonderfully pleasant features to modify and format plain text.
Knap is open source. Over a million people already use Knap directly or indirectly because it started as the templating language I made for Obsidian Web Clipper. Now any tool can use it.
You can add Knap to your app or use it via CLI. Create a batch of Markdown files from JSON or CSV, or pipe Defuddle directly into Knap to generate Markdown from HTML/URLs.
The Knap site is color-coded to help beginners understand the relationship between variables (blue), filters (orange), and logic (green). I've tried to document it as comprehensively as I can.
Try the Knap Playground so you can see how fun the syntax is to use!
Knap is pronounced /knæp/ (with a hard k). It's named after knapping (with a silent k): the process of shaping stones to form arrowheads, scrapers, and other tools.
Enjoy!
