Claude Code и Codex для контента: как настроить систему, чтобы тексты «зашли»
A speaker explains how to train an AI content creation system by manually correcting initial posts and providing feedback to improve output quality. The process involves iterative refinement where the AI learns the user's brand voice and style preferences over time.
Summary
The speaker describes a systematic approach to training AI for content creation, emphasizing that initial posts generated by the system will require significant manual editing and refinement. They explain their personal process of correcting early AI-generated posts, providing specific feedback about language preferences (such as reducing English words and using more human language), and teaching the system to match their writing style. The speaker notes that after approximately 5-10 posts with corrections, the AI system (referred to as 'Code' or 'cloud cloud via telegram') begins to understand the user's preferences and can create content that closely matches their desired style and voice. They emphasize that this method can solve content creation problems permanently, but requires having a clearly defined brand voice first. For those without an established voice, they suggest analyzing content from preferred authors or telegram channels, though they caution this won't truly represent the user's authentic voice. The speaker concludes by promoting downloadable materials and encouraging engagement with their content.
Key Insights
- The speaker claims that initial AI-generated posts will require manual finishing and correction by the user to match their preferred style
- The author argues that after correcting 5-10 posts with specific feedback, AI systems learn to replicate the user's writing patterns and preferences
- The speaker asserts that this training method can solve content creation problems permanently once properly implemented
- The author emphasizes that having a clearly formulated brand voice is a prerequisite for successfully training AI content systems
- The speaker suggests that users can analyze other authors' content from telegram channels as an alternative approach, but warns this won't represent their authentic voice
Topics
Transcript
The first posts that he will send you, they will be there, as they say, not that it may not be very, you can finish them yourself, that is, the first posts that he sent me, I finished them myself, I said, look, I don't like this option, I don't like this option, I fixed it in this way, that is, I manually sat down, corrected it so that this post was as similar as possible to me and told him there, write yourself into the condition that there are fewer English words there and more human language for those options that he threw. You see, there is a huge number of English words here, because we, in my opinion,…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Игорь Зуевич
Сравнение Claude Code, Gemini и Codex — тесты на контент, анализ и изображения
The speaker is comparing AI coding assistants — Claude Code, Gemini, and Codex — by testing their ability to create image-generation agents. Each tool is being evaluated on how well it writes prompts and structures sub-agents for text-to-image and image-to-image tasks.
Экономия токенов и контекстного окна: настройка памяти нейросети – Compact сжатие контекстного окна
The transcript explains the 'Compact' command in AI agent sessions, which compresses the entire conversation history to reduce context window usage without switching sessions. It also covers related commands for full history clearing and automatic triggers for compact execution. The speaker recommends configuring automatic notifications or auto-execution when the context window reaches a certain threshold.
HR-агент на Claude Code: создаёт других сотрудников сам по запросу
The video demonstrates setting up a Telegram HR bot using Claude Code, walking through the process of creating a bot via BotFather, obtaining a Telegram user ID, and letting the code automatically generate the HR agent. The presenter shows how the system initializes itself by installing dependencies and launching the bot with minimal manual configuration.
Живые скиллы для Claude Code, Gemini и Codex | Как настроить super skills для ИИ агентов
The video explains how to create 'living' or 'super' skills for AI agents like Claude Code, Gemini, and Codex that self-update based on user feedback. The presenter demonstrates how these skills collect feedback after each task, store it in log files, and periodically incorporate that feedback into the skill's core instructions. This approach reduces token consumption and ensures skills continuously improve over time.
Планировщик для ИИ агента: 50 идей и готовые файлы для Claude, Codex и Gemini
The speaker describes their use of a scheduling/routing tool called Routins, which they compare to N8N, to automate daily delivery of curated YouTube videos on topics like Gemini and Claude. The system runs autonomously, collecting videos and generating concise summaries of their content without manual intervention.