Browse Comments — Clean (de-noised)
Close reading of the corpus at each pipeline stage: raw → clean → relevant → coded.
47
comments matched
· page 2 of 3
Lists for the sake of lists. Most of them were solved with a couple of paragraphs and a few sentences
Going from one sentence to a fully published video is wild. The fact that this runs on 17 specialized skills and handles everything autonomously shows how far agentic workflows have come. Great build.
Wow serious work this mate well done 👏Claude is seriously the goat 🐐
Frankly speaking I'm really frustrated by the amount of AI-generated content on YT. It is kinda clickbait, you start watching, after a minute or two realizes it is AI and then you stop watching. Simply because of poor quality, glitches and generic inaccurate video of no value.
Most people don’t actually need “autonomous agents,” they need reliable content systems that don’t break when the input changes or the platform shifts. I see this a lot while building Collio AI
Super impressive! What was your learning experience like to understand what Claude code is capable of?
Enrique Marq, I really appreciate the engineering and the idea behind this, it’s impressive work. I do wonder, though: what’s the end goal when this level of automation is applied to social media, which originally existed for human expression and connection? If our social presence is generated by AI, posts written in our “voice,” scripts authored by systems trained to sound like us, what are we actually doing when we “connect” online? At that point, it feels less like people sharing genuine thoughts and reactions, and more like AI systems interacting with each other on our behalf. If everyone is doing this, what’s the point of socializing at all? Part of the value of connection is the time, effort, and intent behind it. Writing something yourself, crafting a thoughtful message, spending attention, that effort is a signal that the people you’re talking to matter. It’s a bit like a handwritten note or choosing a meaningful gift. The effort itself carries meaning. If we outsource that entirely, we may gain efficiency, but we lose the original purpose: the human care behind the act.
Amrit Banerjee moltbook is a fair place for such interactions. The others pretend they are still “authentic”with essentially smart template!
keep building!
We created a “LinkedIn department” that companies can plug into instantly. 25 human-led profiles + 30K monthly outreach touchpoints + 112K existing connections for reputation & reach.
Very strong direction. The biggest opportunity here is not replacing creators, but compressing the time between idea and multi-platform distribution. Whoever builds reliable systems for that will have a serious advantage
Amrit Banerjee
"finds the best moments" Honestly, would not know the best moments if they bit it on its Python track.
Great work my friend 👍
I looked at the comments section here to check how many people tolerate and appreciate being told what to do and how to do it without experimentation and data about real results. And I’ll say this again: Ai doesn’t belong on LinkedIn. 90% of automation connection with the LinkedIn app is being either PUNISHED by the algorithm or BANNING accounts. If not today, then “tomorrow”. Not to mention Ai is thinking for all of us, based on everything it learned about each one of us. And it is using average fractions of the internet. If you saw who was feeding Ai you wouldn’t use it for a shortcut
Thanks for sharing excited to check it out!
And how much would this cost if the "AI" companies were charging at actual cost?
Is Lex Friedman OK with using it’s content?
Spamming on Linkedin. Is anyone interested in bot posts on Social Media?-----------------------------------------------------------------------------------
Enrique Marq this was so well done, and easy to follow, appreciate you sharing!