Mahesh Chulet

Hey @X I'm looking to #connect with people interested in: Frontend Backend Fullstack DevOps Web3 Cloud Freelance Open Source System design Cs Let's grow together 🙌✨

seeing a lot of faceless x accounts talk about "personal brands". that's like a bald man selling shampoo. there's nothing "personal" about your brand if you can't show your face.

the four types of agent loops. loop engineering keeps getting talked about as one thing. it's actually a choice between four structures, and each one fits a different kind of task. it means designing the system that steers the agent, instead of steering it yourself move by move. that system always answers two questions. what starts a run, and what decides the work is done. in a hand-run session you answer both yourself, every single time. each loop type moves more of that into the system. here's each type, what triggers it, and when to reach for it. 1) turn-based. triggered by a user prompt. the agent gathers context, acts, and checks its work inside a single turn, then a human reviews the output and writes the next prompt. use this when requirements are still forming and every output changes what you'd ask for next. 2) goal-based. triggered by a /goal command carrying success criteria and a budget, like "get the homepage Lighthouse score to 90, stop after 5 tries." when the agent tries to stop, an evaluator model checks whether the goal is met, and a no sends it back to work. use this when the outcome is measurable but the path there isn't worth your attention. 3) time-based. triggered by a clock. an interval fires, the agent runs a fixed prompt like "check the PR, fix CI," then waits for the next tick. /loop runs on your machine, /schedule moves it to the cloud so it survives a closed laptop. use this for recurring work where the task is known in advance and only the timing repeats. 4) proactive. triggered by an event or schedule with no human present. a routine watches a channel, and when something needs handling it spawns a workflow with a triage agent, a fix agent, and a reviewer that adversarially judges the work before the task closes. use this for standing responsibilities where you can't predict what will come in, only that something will. each type hands off one more job than the last. turn-based keeps both with the human, goal-based automates the checking, time-based automates the trigger, and proactive automates both while deciding the workflow shape at runtime. so the mapping question isn't which loop is most advanced. it's whether your task is exploratory, measurable, recurring, or standing. the more you hand off, the less you babysit.

Hey @X 👋 I’m looking to #CONNECT with folks interested in: - AI - Vibe Coding - Full Stack Dev - UI / UX - Design - Software Development - Startup - #buildinpublic Let's go together 🚀🤝 #LetsConnect #100DaysOfCode

Hey founders/builders on @X , I would love to #connect with more like minded people who are interested in: • Product development • SaaS • Web development • DevEx • Marketing • AI enthusiasts Let me know what are you working on in the comments below 👇

Satya Nadella may have just written the most influential enterprise AI essay of the year. It also happens to be the best argument against the ecosystem Microsoft benefits from. His core idea is simple: Companies don’t just pay for AI with money. They also pay with institutional knowledge. Every prompt. Every correction. Every “that’s not how we do it here.” Over time, that becomes intelligence the model provider gains from. He’s right. Now ask yourself: Who sits closest to that knowledge? Copilot watches how developers write code. Microsoft 365 Copilot sees emails, documents, meetings, and chats. Azure hosts a huge share of enterprise AI workloads. If enterprise AI has an information leak, Microsoft owns one of the largest pipes. The irony is hard to ignore. The essay criticizes vendors that learn from customer interactions while limiting what customers can do with those same models. But that’s a criticism that lands uncomfortably close to home. Read the piece again through that lens. It’s less a warning about AI. It’s a case for Microsoft’s version of AI. The proposed solution is a “trust boundary.” But if that boundary is built entirely on the same vendor’s infrastructure, how much control have you actually gained? A real trust boundary means your infrastructure. Your model gateway. Your evaluation pipeline. Your audit logs. Components you can replace tomorrow without asking permission from today’s vendor. That’s the difference between owning your AI stack.. and leasing your independence. Curious: When building your AI infrastructure, what’s the first dependency you’d remove if you had to?

Thank you Claude :) Just extend it permanently yeah?

I still can’t figure out why Gemini struggles to compete with Claude and GPT. - Owns Chrome - Backed by Android - Stores most search results - Holds ~95% search history - Google has the biggest user data - Even incognito data isn’t fully private So what’s the problem?

Imagine a fable 5 quality model that’s 3-4x less expensive in less than 6 months. And an Opus 4.8 grade model that can run on a local device in less than 12 months. Greater than 50% chance that these events will happen. Worth keeping in mind when you make predictions about the future.

If you were 20 right now , would you prepare for a job or build a product and become an entrepreneur ?

Hey, founders/builders on @X, I would love to #connect with more creators who are interested in: • Building systems; • Digital Leverage; • Marketing & Sales; • Psychology; • Product Development; • SaaS; 👇 I would be happy to know what you are working on in the comments below.

Looking to #connect with builders who are actually shipping #x 🚀 Founders 🚀SaaS builders 🚀Indie app creators 🚀Growth marketers 🚀Developers Say hi and let’s connect! 😁

You and me when Anthropic extended Fable access for another week