Nate Herk | AI Automation

The video presents nine "tier one" hacks for managing Claude code token usage, focusing on easy-to-implement strategies. Key recommendations include starting fresh conversations between tasks, batching prompts, using plan mode, and actively monitoring usage through various tools.

The video explains 18 token management hacks organized into three tiers to help users overcome Claude's rapidly draining code usage limits. The speaker emphasizes that most users don't need a bigger plan but rather better context hygiene, since Claude rereads entire conversation history with each message, causing exponential cost growth.

Claude Code released a virtual pet feature called /buddy on April 1st that creates Tamagotchi-style companions in your terminal. The pets have different rarities, species, and stats based on your coding history, are permanently tied to your user ID, and provide commentary while you code without affecting token usage.

An analysis of Claude Code's leaked source code reveals it's actually a full agent runtime with sophisticated architecture, not just a chatbot. The leak exposed over 500,000 lines of code showing hidden commands, memory systems, and multi-agent capabilities that most users aren't utilizing.

OpenAI released an official Codex plugin for Claude Code that allows developers to use GPT-4o for code reviews within their Claude workflows. The speaker tested both tools and found they complement each other well - Claude Code excels at planning and creative outputs while Codex is better at code reviews and catching edge cases.

After building hundreds of AI workflows for clients across various industries, the creator identified 5 simple automations that businesses consistently pay for: speed to lead, document processing, follow-up sequences, database reactivation, and internal reporting. These workflows focus on saving time, reducing costs, and eliminating mistakes rather than flashy AI features.

The video demonstrates Paperclip, an open-source tool for creating AI-only companies with multiple autonomous agents. The creator shows how to set up a complete organization with a CEO, engineers, and other specialized agents that can hire each other, complete tasks, and run continuously without human intervention.

Google released Gemini 3.1 Flash Live, a new voice AI model that uses direct speech-to-speech processing instead of converting speech to text first. The model includes visual capabilities, improved performance in noisy environments, and better contextual understanding of emotions like sarcasm and frustration.

Google released Gemini 3.1 Flash Live, a new speech-to-speech voice AI model with improved latency, accuracy, and multimodal capabilities. The presenter demonstrates building voice agents using the model with Google AI Studio and shows how to integrate it with external tools using Claude for coding assistance.