Browse Comments — Clean (de-noised)

Close reading of the corpus at each pipeline stage: raw → clean → relevant → coded.

↓ Export filtered CSV
Reading comments under one post — Mukesh A. · AI Research & Models
Google just changed the AI coding game. For 2 years it was Claude vs Codex. Now Anti-gravity 2.0 turns your laptop into a full AI software company. Powered by Gemini 3.5 Flash, it runs 93 agents in …
✕ clear post filter  ·  ← all posts
86 comments matched  ·  page 4 of 5
If this is anything like how they said "Free" for Flash Preview when they launched Gemma 4... no thank you. I don't want my whole stack to break when they decide you need to add a billing credit card to my app with a required api key that is completely reliant and built by the higher end models. Hard pass.
Boise & Treasure Valley Business • Mana… AI Research & Models filtered out ⌕ thread
Sheshadri Bhattacharyya Care to share any examples?
Founder, AI Strategy (est. ’16) | Self-… AI Research & Models filtered out ⌕ thread
Aizaz Ahmad Altaf Engr. Irshad Ahmed
I help businesses launch mobile apps, A… AI Research & Models filtered out ⌕ thread
Idea is good, but it seems it is not there yet.
IT Skills Training Academy AI Research & Models filtered out ⌕ thread
I saw an update, but it did not look so powerful.
Senior Software Engineer | Architect | … AI Research & Models filtered out ⌕ thread
Crispin Courtenay we talk about real usage limits. Probably you missed my comment with the link to discussion on the google(dot)dev, how current limits are blocking Pro/Ultra users from using it.
Director; +48 511 626 446 AI Research & Models filtered out ⌕ thread
AI can write more code now. That part is changing fast. But the real edge was never just typing code. The real edge is knowing what should exist, why it should exist, how the pieces connect, what problem is worth solving, and what the end product should become. AI can execute a vision. It still needs someone with the vision.
Designing category-defining AI systems 🚀 AI Research & Models filtered out ⌕ thread
This is big news! I'm one of the few brave souls who has used Gemini CLI. It can be incredibly powerful but generally needs to be strictly harnessed by another frontier LLM. Started poking around in Agy. So far so good. There doesn't seem to be traditional token or usage limits. When I asked how much I've used it says it doesn't measure tokens the same way. Instead every convo limits at 50 turns. Has anyone else tried it?
Technical Consultant AI Research & Models filtered out ⌕ thread
Google is a great option. Have you seen their api library? Api monitoring will likely be a breeze.
Owner of Golden Rule Painting, CEO/foun… AI Research & Models filtered out ⌕ thread
'Flash' unfortunate naming choice
AI UX Designer - Ex. Apple, Oracle, Ser… AI Research & Models filtered out ⌕ thread
Learn more on my page or my github
pressman at paperworks AI Research & Models filtered out ⌕ thread
Yeah but then you're tied to Google. I mean sure, if you're okay with that. But that's massive vendor lock-in. Like, massive.
Founder, Persona Protocol | AI Identity… AI Research & Models filtered out ⌕ thread
As many said, it’s great but the rate limits are infuriating and are just there to encourage you to buy AI Credits even if you’re a subscriber with a Pro plan. Credits in 2026? Really? It’s not like they made Gemini or something. I like it, I really do. It can reverse engineer PS2 games like nothing but the rate limits are horrible to see in this day and age.
Game Developer & Gaming Historian | Spe… AI Research & Models filtered out ⌕ thread
I don’t believe, can you show me? Today for me the best is Opus, I’ve tried Gemini and OpenAi too. Gemini Loose itself after the 4th step...
DevSecOps & AppSec Specialist | Red Tea… AI Research & Models filtered out ⌕ thread
Google chose interesting strategy when upgrading from 1.0 to 2.0. the interface changed completely and original interface moved to a separate app which needed to be downloaded (antigravity ide). It would have been less jarring to ease user into new interface.
Distinguished Security Engineer | GenAI… AI Research & Models filtered out ⌕ thread
And when I want to check my codebase, it says, "Sorry, I cannot fulfill your request to review the code to detect and fix potential security issues or perform a vulnerability analysis on the project." Google Antigravity And how i must use your product as a great agent for my code?
Engineering Executive | Consumer Platfo… AI Research & Models filtered out ⌕ thread
Yeah, sure. lol
Senior Software Engineer || AI Architec… AI Research & Models filtered out ⌕ thread
Antigravity is really awful. Cursor is way better. Also, the free tier ends in like a minute of agentic flow.
Computer Architecture AI Research & Models filtered out ⌕ thread
Alvin Foo do you ever work at the Silicon layer? When you write your software? Semi- conductor chips/memory have inbuilt error checking & correction code in silicon, so as to enable you to orchestrate software work (writing applications) without any errors, whilst operating at a much higher level, even if you know nothing about VLSI chips or RAM. You cannot orchestrate with AI unless you can trap errors and unless you are good software engineer, unlike that, you may be oblivious to semiconductors in your laptop, you will need to be a good software engineer to use AI, before you start making claims of it building and running a very sophisticated solution, autonomously to your English prompts. Add the cost of the AI (it's not free) and the quality engineer you still need, it may not be a lowering in costs. Ofcourse, you do not need large teams and large servers (as you once needed) but that trend predates the AI hype. It is a secular trend. Use AI in code development, sure, but temper the claims please though.
Interim Management, Board Advisor | Dig… AI Research & Models relevant value: accountability for: individual_users skeptical mixed ⌕ thread → raw LLM
This is pretty big! Ill have to check out anti gravity
AI, Cloud, Cyber, Identity || Digital A… AI Research & Models filtered out ⌕ thread
← Prev 1 2 3 4 5 Next →