Co-founder @commerce_ui. Running eCommerce AI Lab: we build AI agents for ecommerce brands. DMs open.
Gdańsk
Joined September 2014
- Tweets2.1K
- Following596
- Followers576
- Likes8.4K
super, jeszcze wypuśćmy go za 4 lata za dobre sprawowanie
w polsce jak się chce kogoś zabić to najlepiej przejechać go samochodem
anyone seen "continuous compaction" in the wild?
agent hits any big task that requires exploration → forks itself (same context) → does the work → rewinds, drops the mess, keeps only what mattered → carries on
not talking about narrow subagent but as a 'general' method
Andrzej Dąbrowski retweeted
there is one universal truth:
you have to use emotion for acquisition & substance for retention.
you can apply it to products, dating women, whatever. the same damn principle applies everywhere: what gets someone interested is rarely if ever the same thing that makes them stay.
I've been building ecommerce agents for a while and tried many AI frameworks along the way. I must say:
@eve is the first one that really clicked.
I told @shardara this recently and he asked me why, so I thought it was worth sharing publicly.
In my first AI consulting gigs (mostly custom shopping assistants) I was stunned by how much infra work it takes to ship an agent. I'd say 60–70% of my time went into infra, and only the rest into the actual AI work: context engineering, data, evals. That was insane tbh, because none of that infra felt custom to me. Most frameworks either didn’t help much with infra, or if they did, they forced clunky abstractions on the AI side, and I kept fighting the framework instead of doing my work.
That's probably why so many folks say "don't use a framework, just build from scratch." I felt that too (we all have our Claudes and Codexes, right?), but going framework-less never felt right either. There’s so much repeatable stuff, and way too many footguns, to rebuild it every time.
So for a long time I was looking for something very simple:
1. Pls pls handle infra for me. Durable execution, sandboxes, streaming, reconnects, subagents, file uploads that land in the sandbox, auth, web search, easy deployment.
2. LEAVE ME FREEDOM on the AI part. I want to own the harness, not have the framework own it for me.
@eve is just the first one that ticks all these boxes.
Bonus: every production agent quickly needs a UI and often a custom API. A playground, a context view, a domain-specific data viewer (inspired by @HamelHusain, highly recommend). With Eve + Next.js it’s all in a single repo (not even a monorepo!), deployed with one click. Framework-defined Infrastructure at its finest, it really feels good.
btw, I come from front-end world and I remember exactly how the world looked before and after nextjs. Now it feels like this again, just with AI.
Everything we do at the eCommerce AI Lab at @commerce_ui now runs on Eve: shopping assistants and analytics agents for our clients, plus our own company agent that lives in Slack. It all just works, and a huge amount of custom code just evaporated.
Highly recommend
anyone who understands ecommerce knows it's not gonna play out 100% this way. "the website is dead" was heralded so many times in ecomm: amazon, google shopping, instagram checkout... and shopify stores kept growing anyway!
ofc agents will eat commodity buys
but many brands today sell identity, not just products. and identity only works when it's unfiltered, that's almost the definition of what identity is. the moment an agent summarizes you, you become "specs", the meticulously designed emotional signal is lost. that’s the opposite of desire.
ask yourself: do you prefer to read tweets that feel authentic and "directly-from-human"? or some generated AI summary slop? exactly, and it's the same with many other things than writing
I don't really see any other medium for doing this better than custom website. Pure chatbot as intermediary is surely not an end game here
most of the internet will be completely obsolete very soon.
it already feels like an abstraction we tolerate for lack of a better alternative. ai agents replacing it would essentially reduce the web to a set of apis & scraped data repositories—just raw material for agent-mediated interaction. no need for direct user involvement; you ask, they fetch, synthesize, & act. search engines? obsolete. social platforms? distilled into personalized outputs. even “websites” as a concept might evaporate—why visit a page when your agent brings you the content or makes decisions for you?
the internet becomes invisible infrastructure, like electricity. the downside: agents would consolidate power even harder than platforms like google. imagine them becoming gatekeepers of truth & commerce, & good luck disentangling bias from synthesis. you’d be trading the wild west of the web for a walled garden of agents.
agents will be the de facto mediators of human desire.
idly thinking that humans will always stay curious. nerds are gonna be nerds, and we'll keep having the urge to understand the universe
but if ai understands it 100x better, then I guess it becomes sort of a hobby?
the more complex UI, the more likely it's gonna be eaten by conversational interface
I had to open JIRA today with a bunch of extensions connected. it's so horrible, just give me chat
but when I'm in ecommerce I want nice display of products + quick "add to cart" button
I bet @mati would top the chart if they served a proper beer.
Introducing The Cheeky Pint Benchmark.
We made our intern watch every episode of Cheeky Pint to bring you this breakdown of pints drank by guests to the nearest sixth.
Congratulations to @elonmusk who is currently in the lead with three full pints of Guinness.
It's true, but
1. what happens if app store has 100 similar apps (instead of 2 or 3)? And each one is actually good? doesn't it make everything literally "free"?
2. what if she asks her ChatGPT "what's the best driving test app" and ChatGPT builds it ad hoc?
let me tell you why saas is not dead and how deep in the rabbit hole we are on x
⬖ my wife is technical enough, worked as QA before, understands the concepts of cloud, hosting, frontend vs backend, etc etc.
⬖ she has 2 subs to copen ai and claude, uses them on desktop and on mobile
⬖ she has heard me rambling about ai for 4 years non stop and knows most terms
⬖ she has vibed a few websites for fun and is a creative person overall
so i see she subscribed (recurring lol) to a "learn driving tests" app
i asked her wtf why dont u vibe yourself one?? it will be even better, more visual, tailored to you, etc etc.
she was like "uhm but why, someone already put all the effort and i just wanna be done with my driving tests"
now think about the people who have NONE OF THE bulletpoints above lmfao 💀
put a price tag and just put the saas in the bag lil bro
no one is vibecoding shit
Building customer facing agent is constant fight with internal agent dialogue. The model thinks like a "coding agent", it by default assumes reader understands the tools, terminology etc and it leaks.
It's surprising we don't have any primitive to deal with it out of the box.
Turn any existing Agent into the world's best personal shopper for you.
The new shop skill: shop.md/
Add the skill & connect your Shop account & finally find the thing that you always wanted
it's so true. I've been doing a task for the last 3 days where the first version was made by agent.
it worked, but the critical API area was terrible, I couldn't read it at all, and I had to be able to read it since it was "glue area" between my system and other systems. Duplications, lack of consistency, bloat.
Cleaning it up took me ~3 days. Now I'm thinking that maybe I should have designed this API manually and THEN asking agent to implement it. Maybe it would be actually faster.
YES. This. 10000%.
Your aunt might not vibe code SaaS, but I can.
You built sth cool? Nice, vercel will have a clone in 4 weeks. Can you win with their distribution?
I don't know what's the end game of this, but narrow SaaS with high margins is surely dead.
the fundamental basic economics misunderstanding that ppl have esp on here is that they think normal ppl need to use models to build their own software to be disrupted. they obviously don't & they obviously won’t.
but even if just 1% of the population can build software, that's enough to flood the entire supply side with millions of new products collapsing the cost to near zero.
you absolutely don't need universal adoption to destroy scarcity. you just need enough producers to overwhelm demand. you can feel this now already. i never pay for software because i just ask claude to point to something like a chrome extension or a mac app & it’s just there. at some point your agents will also auto update & maintain your software automatically too.
any time someone from a great quality-driven company has a reasonable and balanced take about ai coding, some random yc kid shows up out of nowhere saying sth like:
SKILL ISSUE DUDE I SHIP 2000 PRs / MONTH THE QUALITY IS STILL INSANE.
if you have good systems you can most definitely ship 100 PRs a day. because you start shipping features at a time
myself and people i know are all accomplished software engineers and this is how we work now
code is not slop if done right. it’s a skill issue but once learnt it’s game changing
we’re just ahead of the curve
People who heavily abuse anything are garbage, but I still don't think robots should have moral status of any sort.
From Asimov 3 Laws of Robotics:
1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2. A robot must obey orders given by humans, except where that conflicts with the First Law.
3. A robot must protect its own existence, as long as that doesn’t conflict with the First or Second Law.
Some extensions to the 3 laws I like:
4. a robot must establish its identity as a robot in all cases
5. a robot must know it is a robot.
It's insane I'm literally thinking about this seriously now
