Building with AI agents in public. Cofounder @TweetsMashApp + LinkedMash(https://xitter.cf/t.co/pbOsWhaJl3). Sharing what actually works.
Planet Earth
Joined June 2020
- Tweets8.6K
- Following521
- Followers17.6K
- Likes38.2K
Claude Code in one terminal, Codex in another, two Claude accounts, and no idea which session was logged in where.
T3 Code put all of it in one window.
→ Claude Code, Codex, Cursor and Grok in one model picker
→ Multiple Claude accounts side by side. Tweetsmash account and personal account, one click apart
→ Every session gets its own git worktree, so parallel agents never touch each other's files
→ Diff, commit, push and open the PR without leaving the app
→ Phone app connects to the session on your Mac, so the agent keeps working when you step away
Open source. It wraps the CLIs you already use, so no new subscription.
If you're like me and recently got Claude startup credits on a new account, but still want to use your existing Claude plan, this is the easiest way to run both without logging in and out.
The catch: it is still pre-1.0. Expect rough edges.
The real upgrade is not a better model. It is one place to run all of them.
t3.codes
Never deleting this app.
7 days stuck in Play Store review.
31 minutes to get a reply from Google's CEO.
The fastest support ticket in tech is still a public post.
AI answers are too long to read. Stop asking for text.
Karpathy's fix is a ladder of output formats, each one easier to absorb than the last:
1. Strict English. Ask for "80% of the way to ASD-STE100", the writing spec for aerospace maintenance manuals. About 900 approved words, max 20 words per instruction, active voice only. Dense answers become readable.
2. A diagram. Ask for a picture instead of paragraphs. You see what connects to what in seconds.
3. A web page. Ask for the answer "in HTML". You get something you can click, filter, copy from and share.
4. An explainer video. "Create a 3b1b style video explainer on X." Narrated, animated, made for an audience of one.
5. The real shift: as agents do more of the work, your job moves to reviewing and understanding it. Custom throwaway apps and videos are now cheap enough to build for a single question.
Two of these already run my week with Chiti:
-> Anything I paste into another tool arrives as an HTML page with a Copy button. Copying from the terminal once dropped characters into a live post.
-> Analysis-sized answers come as a narrated video. macOS voice, headless Chrome, ffmpeg. $0 for audio, about 2.8k tokens to script 14 slides.
Reading more was never the bottleneck. The format was.
We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks:
Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better:
Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better:
Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better:
Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work!
In summary:
- As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding.
- Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.
AI agents are checking your pricing page before your buyers do. Your analytics never shows it.
Here's how to catch them and fix it:
(Analysed and learned from my products)
1. open your analytics or server logs and pull the last 30 days
2. filter by user agent: GPTBot, ChatGPT-User, ClaudeBot, Claude-User, PerplexityBot, meta-externalagent
3. export as CSV with 3 columns: path, user agent, date
4. give the CSV to Claude with this prompt:
"Group these AI agent visits by page. Show the 5 pages agents read most, which agent read each one, and what question a buyer was most likely asking when the agent landed there."
5. if /pricing is in the top 5, open it and check: is every price, limit and plan plain HTML text? Not an image, not a toggle that needs JavaScript, not "contact us".
6. then this prompt:
"Read [YOUR PRICING URL] as an AI agent comparing tools for a buyer. List every fact you could not extract with confidence: prices, limits, what is free, what is paid. Order them by how likely each one costs me the recommendation."
7. fix every item on that list
8. open Bing Webmaster Tools and click AI Performance, not Search Performance. Ours showed 52 clicks in 3 months on Search. AI Performance showed 156 Copilot citations in the same window. Most founders only ever open the first tab.
9. ask Copilot for the best tool in your category and look at what it cites. For us it was Chrome Web Store and Edge Add-ons listings, not our site. Our Chrome listing was set to Unlisted, so it served noindex. One toggle fixed it.
10. resubmit the fixed pages: Google Search Console, IndexNow for Bing, and search.brave.com/submit-url for Brave (the /webmaster URL most guides name is a 404)
Agents skip your homepage story. They read your pricing table. Make it legible to a machine before you make it pretty.
Just asked Opus 5.5 to make a video on the history of computers.
Educational industry should be paying attention.
Why do Claude and ChatGPT both cost $20 a month?
Asked Opus 5.5 to explain it with Nash equilibrium.
First it used the prisoner's dilemma. Correct, but I still didn't get it.
So I changed the ask: explain it like I'm 5, with things I already know.
The version that worked: 2 ice cream carts on a beach, then the $20 price tag. Took 3 tries. Clicked in 2.5 minutes.
The prompt, for any concept:
"Make a 2-3 minute whiteboard video explaining [concept]. Explain it like I'm 5. Start with one everyday example I can picture, then connect it to something current I already know."
Opus 5.5 + TweetsMash = 🔥.
Opus made this product video on a real product loop:
lost X bookmark → library → sync → search
End line: Your X bookmarks, finally useful.
How it was made:
- HTML pages as frames
- Real bookmarks pulled from the API
- Headless Chrome capture
- ffmpeg stitch
- Python music track
- 8 renders after taste feedback
39 seconds. One core loop.
Opus is strong at launch video when the brief is cruel:
one job, real UI, hard rules.
Steal-this prompt in the reply. tweetsmash.com
Loving this little app❤️
I was looking for something to brain dump thoughts, track tasks, and focus on one thing at a time, without digging through windows.
Accidentally stumbled on NotchOwl and switched immediately.
Mac users, give it a try. One-time $4.90.
Worth experimenting.
$16,000+ earned from @linkedmash since launch $1,614 of it in the last 71 days.
Every number and every mistake below.
The public goal was $2,000 MRR. The chart stopped at $363 and the challenge stopped on day 65.
The money
30 payments from 27 customers, $1,614 collected before Stripe fees. July $722 net, August $748, September $61 so far. 32 of 37 subscriptions were annual, so cash landed in $84 and $120 lumps while the MRR chart moved $7 and $10 at a time.
The misses
$300 by 31 July, missed at $249.50. $550 by 14 August, missed. $400 by 31 August, missed at $349.15. Required pace by the end was $39 a day. Best pace ever run was $5. 38 of the first 53 days were flat.
The mistakes
1. Built instead of sold. Discover, an onboarding wizard, an Agency tier, a $49 Exports Pass, a profile analyzer, a digest. Agency deals 0. Exports Pass buyers 0.
2. Yearly checkout by default. A $120 sale books $10 on an MRR chart. Wrong scoreboard for a public MRR goal.
3. Nobody searched. 13 of 17 signups in one 8-day stretch imported 13,665 posts and never ran a search. Zero reached the paywall.
4. Sold one job as a subscription. 5 same-day cancels, the fastest 18 minutes 45 seconds after a 2,062-post export. Removed the $14 plan on day 62.
5. No lifecycle emails for weeks. One customer hit 9 failed $7 charges and was never told.
6. 12 accounts made 94% of all API calls and paid $45 a month between them. Charging them was the only lever that cleared $2,000. It stayed a plan.
7. 4 posts went out with an MRR number $19 too high because the previous day was reused instead of pulling Stripe.
What worked
- 388 comeback emails to old signups: 9 clicks, 1 annual customer in 4 hours.
- 536 launch emails: the first 3 replies found 3 bugs. Talking to people who already used the product beat every feature.
If starting again
- Set the target from the base, not a round number.
- $19 minimum, monthly default, from day 1.
- Charge the heaviest API users in week 1.
- Force import to first search before any new feature.
- Emails on before posting anything.
- Focus more on distribution than building.
62 posts in 71 days, 3 stalls. Around day 55 the post became the work instead of the receipt of the work. LinkedMash keeps running. The upcoming days focus goes to TweetsMash and new products.
Here is the link for trustmrr of linkedmash - trustmrr.com/startup/linkedm…
Cool, @linkedmash got 12 points on producthunt.
Day 65/100 of taking linkedmash to 2k mrr. Current - $363
I'm publicly ending my challenge here, and focusing on my products @TweetsMashApp and upcoming new products.
Learned lots of valuable lessons. Thanks everyone for the support.
We're live on Product Hunt.
LinkedMash - your LinkedIn saves, searchable and ready to post with support for agents.
Try it and tell us what's off:
producthunt.com/products/lin…
Same founders as Tweetsmash.
New Premium today - +$19 MRR.
Also shipped two quiet fixes:
• DataFast was double-logging some payments (Stripe was fine; analytics wasn’t)
• LinkedIn avatars now rehost so identify works instead of 403 cartoons
Small trust work while we keep building toward $2k.
Day 63 of 100, taking LinkedMash to $2k MRR. Current $382.15 (+$19)
