Andrei Sheina

@dogukanbuilds @ClaudeDevs This is the way!

@AndreiSheina @ClaudeDevs Well, that's right :D I definitely learnt that if I go rogue with Fable 5, it will write 25 MD documents in the repo that our lead dev will spend 5 hours to debug and remove ahaha

How do teams get agents into production? New blog post from our Applied AI team on Claude Managed Agents and the challenges it solves (credentials, sandboxing, observability, & more) ...

@dogukanbuilds @ClaudeDevs You learn from each fck up thought :) Reduces the probability to fck up on the next deployment!

@ClaudeDevs What we actually do is, > introduce the agent into production > agent fucks everything up > dev is now angry with me > he introduces the agent But this is great lol

@ThePeterMick agentplace.io Build, sell, and scale agents for Claude Code, Cursor, Codex, and beyond 😎

drop your startup link

@e2e_developer Don't you think that AI will get better an better with code over time thanks to A players helping to train it? At the end of the day if mediocre code works and does it's job, does it really matter that it's mediocre? Given it's secure of coarse

tl;dr learn how to build software I want to be frank. AI writes mediocre code. It's flawed. But, from what I know by experience, it's the same mediocrity you find out there. I am fascinated by AI because it's not worse of what I've seen in usual software projects. The truth is we have more C players than A players around. We need to make sure that we implement practices, so that AI doesn't get away with writing bad code. I am doing a QA sprint for one of my projects at the moment. The good thing is, that AI often sees that something is off. But it often doesn't see the issue. That's where a senior operator is needed. So, you don't need to learn DSA to become good with agents. But you need to learn how to continuously deliver software and how to keep it from breaking

@e2e_developer @_EvgenySorokin @Agentplace_io Haha, it looks like it’s actually alive. Turned out Cyberpunk is not 2077, it’s 2026 😅

@_EvgenySorokin @Agentplace_io Looks very cool. Maybe a bit too cyberpunk-esque for some people.

Hey builders, I've laid my hands on UI again (deep down I know I'm not a designer) and want your honest opinion on the realtime audio feature visuals for @Agentplace_io platform. Please look at the animation - what do you think? Is it too much?

That’s a good idea. The way it’s currently implemented - with credits being deducted directly from Agentplace account, it’s already quite safe - you won’t be charged extra if you hit your credits limit. But the ability to be even more flexible and fully control your spending is really good 👍

@AndreiSheina I guess it would be cool to let users to setup budgets and automate this process even more

Agents now evaluate what they need and buy it - you just approve. Here's one building a competitive research tool from a single prompt 👇

Try it yourself at agentplace.io

@Agentplace_io @Systerr Agentic economy is already here

@Systerr Finally, I have money to spend, let's shop..

Recently at @Agentplace_io we've been thinking about how to extend agent functionality beyond what our subscription model covers. What if the AI agent itself (our editor) could purchase additional services to solve user problems?

Agents can now buy tools they need to code with your approval (for now)

@kaysin24343 Nice! Will definitely give it a try

@AndreiSheina No special prompt or template needed, you can just tell the builder what you want, e.g. "build an assistant agent for my team with GitHub, Langfuse, etc. integrations" and it'll set the whole thing up for you 🙂

Publish your agent and connect it to Claude Code in seconds.

Anthropic is keeping focus on text is exactly what makes them the best for reasoning, writing and code. IMO OpenAI trying to have it all backfired for them (they’ve closed Sora), Google just has so much cash to burn that they don’t care and Grok… Have you actually used Grok for anything? 🙂

Claude is arguably the best AI for reasoning, writing, and code. But in 2026, it still can't generate a single image or video. OpenAI has it. Google has it. Grok has it. Is Anthropic making the smartest long-term bet or leaving a massive gap? What do you think?

Been building AI agents in both Claude Code and @agentplace_io lately. In Agentplace there's a dedicated "Connections" tab — you add Gmail, Notion, GitHub, etc. once, and reuse them across every agent you build. In Claude Code, you're configuring MCP servers per project (or globally via --scope user), managing JSON configs and OAuth flows. Reusing them is a bit painful. Maybe I'm missing something — is there a better way to handle third-party connections in Claude Code?

I recreated Dan’s Codex-native agentic app in 15 mins.