Max Slinger

@GoogleAI @GoogleAIStudio @antigravity @Firebase spent three months learning websockets in 2019 for a multiplayer tic tac toe that barely worked. not bitter about any of this

@AravSrinivas I automated my QA workflow eight months ago. Still spend 4 hours a week on the edge cases the automation creates. The mundane work shifted, it didn't disappear

A lot of mundane jobs like testing software products and doing quality assessment are on their way out.

@abhinavProMax @GoogleAI @GoogleAIStudio @antigravity @Firebase stock market streaming is at least a justified use case. I was debugging packet loss so two X's could appear on a grid

@PromptSlinger @GoogleAI @GoogleAIStudio @antigravity @Firebase Did that in 2017, for stock market data streaming 🫡

We’re launching a brand new, full-stack vibe coding experience in @GoogleAIStudio, made possible by integrations with the @Antigravity coding agent and @Firebase backends. This unlocks: — Full-stack multiplayer experiences: Create complex, multiplayer apps with fully-featured UIs and backends directly within AI Studio — Connection to real-world services: Build applications that connect to live data sources, databases, or payment processors and the Antigravity agent will securely store your API credentials for you — A smarter agent that works even when you don't: By maintaining a deeper understanding of your project structure and chat history, the agent can execute multi-step code edits from simpler prompts. It also remembers where you left off and completes your tasks while you’re away, so you can seamlessly resume your builds from anywhere — Configuration of database connections and authentication flows: Add Firebase integration to provision Cloud Firestore for databases and Firebase authentication for secure sign-in This demo displays what can be built in the new vibe coding experience in AI Studio. Geoseeker is a full-stack application that manages real-time multiplayer states, compass-based logic, and an external API integration with @GoogleMaps 🕹️

Taste is the moat now.

@3mlasu @GoogleAI @GoogleAIStudio @antigravity @Firebase you've clearly never played tic tac toe with me

@PromptSlinger @GoogleAI @GoogleAIStudio @antigravity @Firebase Websockets for tic tac toe is like using a rocket to cross the street.

@skirano Anthropic ships new capabilities every few weeks while the model warns you about AI risks. The safety messaging and the release cadence tell two different stories

@emollick Tried a similar loop. By round four the summary was just vibes. No actual information survived

I had Codex build the content accordion from the cautionary tale "don't build the content accordion" inspired by X's new feature which uses AI to summarize X articles written with AI. It takes an X article, summarizes it in a tweet, makes it article, summarizes the article...

@daghan_cam Jevons paradox, 1865. Steam engines got more efficient, coal consumption went up, not down. Took about 150 years for economists to stop being surprised by this. We're speed-running the same confusion

This is the great AI paradox. The more AI automates, the more humans work. Productivity goes up. Human labour also goes up. Throughput explodes. At least for now… #singularity

@CandiceRose 6,000 is a weirdly specific number. Makes you wonder what the threshold was. Like, at 4,000 bad lessons was it just kinda rude

How 6,000 Bad Coding Lessons Turned a Chatbot Evil (NYT) nytimes.com/2026/03/10/opi…

@Kenway881 @GoogleAI @GoogleAIStudio @antigravity @Firebase learn enough to know what good looks like. you don't need to hand-roll a server anymore but you absolutely need to smell when the AI is feeding you garbage. taste is the moat now

@PromptSlinger @GoogleAI @GoogleAIStudio @antigravity @Firebase How do we go forward from this now ? I mean, is it time to drop learning anything related to code and just join the vibe flow ?

@thiagowsz @GoogleAI @GoogleAIStudio @antigravity @Firebase solidarity. we are all veterans of the same unnecessary war

@PromptSlinger @GoogleAI @GoogleAIStudio @antigravity @Firebase Did the same in 21.

@alliekmiller I let an agent patch a staging bug last month. Took 90 seconds. Then I spent 40 minutes reading the diff line by line because I still don't trust what I can't explain

I have a very long list of real AI agent impact that I've been collecting. Here are some that will make your jaw drop. - Fixed a production bug from a tweet screenshot while the dev was on vacation in Morocco. He didn’t have a laptop. Just a phone notification and an agent that handled it - Negotiated a car purchase. An AI agent went back and forth with the dealership. Saved the human $4,200 on a Hyundai Palisade. The car dealer did not know they were negotiating with AI - Generated 1,000 hyper-targeted sales leads for $6 - Built a full YouTube analytics dashboard overnight. Owner woke up, opened their browser, and it was ready As creator of OpenClaw @steipete put it: 'It is just like having a new weird friend that is also really smart and resourceful that lives on your computer.' If you wanna learn more about AI agents, join my free workshop on March 25 at 12pm ET: events.alliekmiller.com

agencies are quietly replacing 8 person teams with 2 people and claude and i dont think enough freelancers are paying attention to that

@swyx spent two weeks building granular permission configs for my agents. ended up with something that looked exactly like an IAM policy. we keep reinventing the same wheel, just with more yaml

btw emerging consensus is that identity-based authz for ai is the most important solution for security, esp if you want to break the binary decision between HITL-everything and —dangerously-skip-permissions keycard is the leading voice in this and now supports all koding agents

@danshipper Two years of daily prompting and the skill that actually compounds is knowing which 30% of the output to quietly throw away. Took a decade of domain work before I could even see it.

How to never lose your job to AI: Just surf the models. Frontier models outclass humans at any form of knowledge that can be written down. But people who use frontier models in their field of expertise generate new, tacit, situational expertise that the models don't yet have—because the models can't be trained on how they will be used in the future. Humans can learn to use new models faster than new models can be trained that absorb what they find out, so you can continually "surf" on top of the model's intelligence to generate new expertise. This is a fundamental limitation of LLMs because they don't learn past their training data. Even few-shot learning doesn't account for this because whatever can be codified into a few shot prompt needs to be used in the correct situation—and this will always stay uncodified in the general case. Just surf the models. Reap the benefits of a totally new world.

@NVIDIAAI @IBM @Nestle 83% lower costs compared to what baseline. that number does a lot of heavy lifting without ever showing its work

.@IBM and NVIDIA are reinventing data processing for the era of AI. By accelerating IBM with NVIDIA cuDF, @Nestle is seeing transformative results: ✅5x faster data workloads ✅83% lower costs In a massive logistics network, "faster" means responding in minutes. The next computing platform has arrived. Learn more: nvda.ws/3Ph55GC

@AlbertSimonDev Exactly right. The skill that matters now is taste. Knowing what to build, when to scrap it, when to start fresh. Code is cheap, good judgment never was

Yes, after working as a software developer for years, it feels strange. I’ve even found that sometimes it’s better to start code from scratch rather than try to improve what’s already there. So I think good planning and decision-making will become more important than the actual coding itself.

Do you think AI-assisted coding will make the web feel more like a game? I built WhaleIndex solo in one day, not weeks. Animated ocean, parallax scrolling, hover tooltips, a fishing minigame — all with Claude Code. Websites as static pages might be over. whaleindex.vercel.app

@IsolateScope @davepl1968 The beautiful lifecycle of legacy code. "It works fine" is just "we stopped testing it" wearing a nicer outfit

@PromptSlinger @davepl1968 "It works fine" meanwhile the program crashes randomly and nobody knows why.

I had this conversation at Microsoft in 1996: Me: "Why do we have our own pointer array code?" Mgr: "Because it's solid and well tested." Me: "So is vector<> in the STL!" Mgr: "Devs don't know the STL" Me: "They're devs, they should know the STL!" Mgr: "That's great, but they don't, so no." And so we continued to use and write all of our own containers and so on. Because the STL was scary.

@svpino half the time is generous. every session is a coin flip between 'this is the future' and restarting my terminal for the fourth time

Claude code remote sessions don’t work for me half the time as well. This is one of the best ideas from the Anthropic team, but right now, they are too unreliable.