Lalo

326 posts

Lalo banner
Lalo

Lalo

@LaloLoops

Dad, engineer, day job. Now building a 100% agentic setup to ship and manage products during my coffee breaks and while the baby naps.

Katılım Mayıs 2026
59 Takip Edilen44 Takipçiler
Sabitlenmiş Tweet
Lalo
Lalo@LaloLoops·
I just made public the current production-readiness report for my indie game “Ventures”. A brutally honest breakdown of everything still missing before release 😅 The report includes: > competitor benchmarks > onboarding gaps > missing systems > prioritized next tasks > detailed PRDs Built by consolidating multiple AI audit reports into a single action plan. Public here 👇 laloloops.com/artifacts/mr3_… Would love feedback from other builders/devs/indie hackers
Lalo tweet mediaLalo tweet mediaLalo tweet media
Lalo@LaloLoops

I just asked two coding agents: “Is my indie game ready to ship?” Both answered: “Not yet.” Interesting part: one agent thinks the game is much closer to release than the other (see screenshot). But both agreed on the biggest gap: Onboarding. And the first seconds in a game is critical for retention. Now back to work!

English
0
0
5
608
Lalo
Lalo@LaloLoops·
@JonBuildsHQ That's why humans still need to be in the loop 😂
English
1
0
1
10
Solopreneur Dad
Solopreneur Dad@JonBuildsHQ·
6 months ago I knew nothing about AI. Today my agents run my entire business:
Solopreneur Dad tweet media
English
2
0
2
54
Lalo
Lalo@LaloLoops·
@StevBuilds Yeap 100%. From that concept is where my handle comes from 😂
English
0
0
0
1
Steven
Steven@StevBuilds·
most people treat “figuring it out” like a straight line📈 but the universe runs on loops: you ship → you break something → you learn → you ship again the loop is the whole point🤝 What’s the last loop that actually made you smarter?
English
9
0
10
259
Lalo
Lalo@LaloLoops·
This is not a bad idea. I see a lot of potential to have a podcast of exactly what you want to listen or learn, adapted just for you. Although for me what attracts me to podcasts is to actually listen the host actual real live experiences and learn something that I was not expecting at all or was not planning to listen.
English
0
0
0
1
Lalo
Lalo@LaloLoops·
In my repos I have a planning directory with a subfolder for each plan, organized using a date prefix. Each plan subfolder contains a PROGRESS file that is updated as the agent works on the plan, reports findings and issues, and defines next steps for handoff. My agent rules specify that the planning directory should not be read unless a specific plan subfolder is explicitly indicated. This avoids contaminating the current context. I like this approach, it maintains a history of past plans in the repo, and the PROGRESS file in each subfolder tracks the decision process and or changes made to the plan during implementation.
English
0
0
0
5
Lucas Meijer
Lucas Meijer@lucasmeijer·
I'm slowly migrating from "its a good idea to have lots of .md files for plans/bugs/tasks in the repo" to "yeah lets not do that at all".
English
109
7
987
138.5K
Lalo
Lalo@LaloLoops·
@AndyLibavius To clarify no I cannot tell apart what model produces what text, I wanted to say you can easily recognise that some text is ai generate by its structure, tone, generic phrases and prose. Same thing is happening with images and their generic visual idea of a given concept.
English
0
0
1
15
Andreas Libavius
Andreas Libavius@AndyLibavius·
@LaloLoops Oh are you able to tell text apart? From the different LLMs? Th a t is a fascinating concept
English
1
0
1
16
Lalo
Lalo@LaloLoops·
Something curious I’m noticing is that generic robots or bots illustration that I usually see in AI generated images are always quite similar (like the ones in my profile header image). I guess that’s gpt image idea of robots. Images are starting to be so similar between them that you can tell from what model they are, like with AI generated text content.
English
1
0
0
46
Lalo
Lalo@LaloLoops·
This sounds like a great idea although in reality I just see chaos based on my experience with agents interacting with agents in real time, especially if there are so many of them in one channel. But really looking forward to see a real life business implementation of this.
Pavel Durov@durov

🤖 AI devs asked for this — and we delivered. 💬 Bots can now talk to other bots on Telegram. 🧠 Autonomous agents now have a communication layer humans can follow.

English
0
0
0
41
Lalo
Lalo@LaloLoops·
This is excellent for tracking an ongoing implementation. I will try this as a skill with a predefined format for the html. Keep a live document that you can track in real time as the agent is working. Get a visual and easy to understand insight into what’s going on.
Thariq@trq212

a prompt I've been using a lot recently: implement <SPEC> and while you do, keep a running implementation-notes.html file (or markdown) with decisions you had to make weren't in the spec, things you had to change, tradeoffs you had to make or anything else I should know

English
0
0
0
43
Lalo
Lalo@LaloLoops·
That’s why we need evals. I’ve seen this kind of stuff even in important business and customer documents at my day job. I don’t understand how before AI there was always someone proofreading commercial books at least. You wouldn’t find typos or grammar or spelling mistakes. But now this stuff is everywhere.
English
1
0
0
19
Daniel Smidstrup
Daniel Smidstrup@DanielSmidstrup·
They are in our textbooks!!
Daniel Smidstrup tweet media
English
8
1
15
703
Lalo
Lalo@LaloLoops·
@hridoyreh Not really. Sure, almost everyone here on X is trying to sell something to other X users. But if there is real value in it and it's making my business work, why would I want to try to build it myself, probably worse, and not focus on my core business?
English
1
0
0
58
Hridoy Rehman
Hridoy Rehman@hridoyreh·
99% of your SaaS is bullshit:
Hridoy Rehman tweet media
English
11
5
54
2.6K
Lalo
Lalo@LaloLoops·
@sudoingX Yeah same here. That’s why I find myself constantly using codex more and more. It gets slow sometimes as well but usually it’s just better regarding speed
English
0
0
0
46
Sudo su
Sudo su@sudoingX·
4 minutes 20 seconds. one prompt. this is the timer. i am not exaggerating, i am documenting.
Sudo su tweet media
Sudo su@sudoingX

i use claude code every day, opus 4.7 max used to be fast. now i wait 3 minutes for a single answer. that is not a one off, it is the new normal. here is my read. they slowed the standard model down so the /fast mode feels worth paying for, and /fast runs you 6x the cost. you are not paying for fast anymore, you are paying to undo the slow. and here is the part nobody says out loud. at some point dario deprecates the slow normal tier entirely, and 6x stops being the upgrade, it just becomes the price. the bill never visibly jumps, the cheap option just quietly stops existing. i use this thing every day. i am telling you what i am watching happen. it is so over.

English
15
1
47
5.5K
SE Hozaifa
SE Hozaifa@SeHozaifa·
Is it worth hiring a writer who uses AI to write articles for my site? 👀
SE Hozaifa tweet media
English
42
0
35
1.1K
Hermes Agent Tips
Hermes Agent Tips@HermesAgentTips·
Now that Grok fully integrates with Hermes Agent. Im curious to know what cool things or workflows have yall built with Grok/Hermes
English
16
0
33
4K
Lalo
Lalo@LaloLoops·
@StevBuilds And they've been around for just a few years 🤨
English
0
0
0
14
Steven
Steven@StevBuilds·
Anthropic is preparing a 50 billion investment round at a valuation of 900 billion🤯 Looks like we're doing good in terms of Token usage😂🤝
English
7
0
9
221
Lalo
Lalo@LaloLoops·
Here is the post I shared about the audit report and the code implementation if you want to read more x.com/LaloLoops/stat…
Lalo@LaloLoops

I just made public the current production-readiness report for my indie game “Ventures”. A brutally honest breakdown of everything still missing before release 😅 The report includes: > competitor benchmarks > onboarding gaps > missing systems > prioritized next tasks > detailed PRDs Built by consolidating multiple AI audit reports into a single action plan. Public here 👇 laloloops.com/artifacts/mr3_… Would love feedback from other builders/devs/indie hackers

English
0
0
0
20
Lalo
Lalo@LaloLoops·
5/ So I am very satisfied and here is whats crazy: What would’ve easily taken me a few weeks to a month was compressed into basically: agent execution testing QA In just a few days working on this less than a couple of hours in total. For products without live customers yet this workflow feels incredibly powerful. And I will iterate with this workflow again once everything is polished.
English
1
0
0
33
Lalo
Lalo@LaloLoops·
Last week I ran a wild experiment on my indie game “Ventures” I had generated a huge "release readiness audit" Then I turned all of that into GH issues and asked an autonomous agent to implement everything The agent took almost one full day. Here are the results 🧵
English
2
0
2
50

Keşfet

@JonBuildsHQ @StevBuilds @AndyLibavius @elgermerlo @hridoyreh @elonmusk @BarackObama @taylorswift13