Lobster Father: New Telegram AI Agent is INSANE!
Telegram's April 2026 infrastructure update introduced a system allowing one AI agent to create and manage other AI agents, with Lobster Father being an early no-code tool built on this infrastructure. Users can create a master AI agent in two taps that routes tasks to sub-agents, integrates with Telegram Business, and automates community management without any coding knowledge.
Summary
The video covers Telegram's major April 2026 infrastructure update, which introduced native AI agent capabilities to the platform's approximately 1 billion monthly active users. Telegram founder Pavel Durov announced on April 15th that users can now spin up a personalized AI agent in just two taps, a dramatic reduction in the setup friction that previously required technical knowledge and caused most people to abandon the process.
Lobster Father is highlighted as one of the first no-code tools built on top of this new infrastructure. The workflow involves interacting with Lobster Father inside Telegram to receive a token, which is then connected to third-party platforms like Telawware, GPTA, or Lazy AI. From these platforms, users can configure their agent's conversation style, assigned functions, and underlying AI model (GPT or LLaMA) entirely through menu selections — no coding required.
A key feature is the master agent system, where one top-level AI agent can manage multiple sub-agents, each assigned to specific tasks or Telegram groups. This is particularly relevant for Telegram Business users, as the agent can autonomously reply to messages, welcome new members, filter spam, and answer FAQs around the clock. The update also introduces agent-to-agent communication, enabling complex multi-step automated workflows that previously required custom development.
The presenter cautions against overselling the technology, noting that while the infrastructure and tools are real and confirmed in Telegram's official API documentation, the actual value depends entirely on thoughtful configuration — identifying which tasks are genuinely repetitive and low-judgment before deploying automation.
The video also covers other features in the April 2026 update: an AI text editor built into the message bar that offers grammar fixes, translations, and style rewrites (including formal, Viking, and biblical styles) with privacy guarantees via Telegram's Cocoon network; Mighty Polls; live photo support; and an iOS document scanner with multi-image stitching and PDF conversion. A new warning label system for accounts using unofficial third-party apps was also noted.
Key Insights
- Pavel Durov announced on April 15th that Telegram's new infrastructure allows one AI agent to create and fully control other AI agents on behalf of users, reducing the entire setup process to two taps — a dramatic shift from the previously clunky, technically demanding manual process.
- Lobster Father enables a master agent architecture where a top-level AI agent automatically routes tasks and messages to specialized sub-agents, allowing users managing multiple Telegram communities or inquiry types to automate coordination without manually jumping between chats.
- Telegram's new agent-to-agent communication support allows AI agents to directly hand off tasks to other AI agents, enabling multi-step automated workflows that previously required custom development to build.
- The presenter argues that the real skill in using these AI agent tools is not simply setting them up, but thinking carefully about which tasks are genuinely repetitive and low-judgment — because an agent connected to the wrong platform or powered by a mismatched model will not save time.
- Telegram's new AI text editor, which appears after typing more than three lines, offers style rewrites including formal, short, tribal, corporate, zen, biblical, and Viking — and Telegram claims all processing happens in a private environment with zero access to personal data via their Cocoon network.
Topics
Transcript
[0:00] lobster father new telegram AI agent is insane. You're still doing everything manually. Setting up tools, copy pasting tokens, figuring out which platform does what. And while you're stuck doing that, other people are automating their entire workflow in two taps. Not 2 hours, two taps. That's what just happened on Telegram. And if you haven't heard about Lobster Father yet, you're already behind. Stay with me because by the end of this video, you're going to understand exactly what this thing does, how to use it, and why it actually matters for anyone trying to work smarter with AI. Hey, I'm the digital avatar of Julian Goldie and I help people learn and actually use AI tools [0:31]…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Julian Goldie SEO
NEW Nvidia Autonomous AI is WILD!! 🤯
Nvidia announced Nemo Clo, a new autonomous AI agent system that operates independently without continuous prompting. Powered by Nemotron 3 Ultra (a 550 billion parameter model), the system is five times faster and cheaper than previous versions, with OpenShell providing secure sandboxed execution.
Laguna XS 2.1: New FREE + Opensource Local AI!
Julian reviews Laguna XS 2.1, a new free open-source local AI coding model from Poolside that performs comparably to Qwen 3.6 and outperforms Claude Haiku on benchmarks. He demonstrates its practical capabilities by building landing pages and functional apps, highlighting its speed, offline functionality, and multiple deployment options through local setup, Claude Code, or OpenRouter's free API.
How to Run Hermes FREE Forever!
The video demonstrates how to run the Hermes AI agent for free using Gemma 4, a local open-source model from Google, with significant speed improvements through MLX optimization. The setup works on Apple Silicon Macs or via free APIs on Open Router, enabling autonomous agents to work offline and privately without subscription costs.
This NEW Chinese AI is INSANE! (FREE + Open Source!)
Long Cap 2.0 is a new open-source Chinese AI model from a food delivery app company that offers 1 million tokens of free context memory, beats GPT-4.5 on SWE bench pro benchmarks, and uses efficient parameter activation to reduce computational overhead while maintaining high performance.
Claude Code is now FREE: Here’s how…
Google's new Gemma 4 model running on Ollama is 90% faster on Apple Silicon, enabling free Claude Code usage locally without token costs. The setup requires three simple steps: downloading Ollama, Gemma 4, and installing into Claude Code, with alternatives available via OpenRouter API for non-Mac users.