Creating AI Videos with VEO3 on Autopilot
Today's curated insights on AI product management from X/Twitter across 60+ expert sources and YouTube channels.
Creating AI Videos with VEO3 on Autopilot
From X
AI Product Launches & Updates
AI Studio & Gemini launch: Logan Kilpatrick @OfficialLoganK announced Google’s AI Studio and Gemini to accelerate product development and gave an early preview of the AI Studio home page with dynamic context switching.
Perplexity Comet Enhancements: Arav Srinivas @AravSrinivas teased new Perplexity Comet features to streamline legacy website interactions and an agentic browser for mobile natural-language task dictation.
AI Tools & Applications
Deep Research Agent with LangGraph: LangChain AI @LangChainAI detailed how Exa built a production-ready deep research agent using LangGraph to deliver structured web answers for complex queries.
Multi-Modal Podcast Researcher: Philipp Schmid @_philschmid showcased a Multi-Modal Podcast Researcher that transforms YouTube videos and research topics into full reports and podcasts using Gemini 2.5 and LangGraph.
LlamaCloud MCP Server OSS: LlamaIndex @llama_index open-sourced the LlamaCloud MCP server for connecting private data and LlamaExtract to MCP clients like Anthropic’s Claude Desktop.
Product Management Insights & Strategies
Combating Lazy Assumptions: Shreyas Doshi @shreyas warned against “lazy thinking” in talent evaluation, noting how assumptions like “you did X, so you must be Y” can mislead product leadership hires.
Navigating Ambiguity in Careers: Shreyas Doshi @shreyas explained that our need to eliminate ambiguity & uncertainty often undermines authentic career decisions, causing repeated mistakes.
Mastering Analytical Thinking Interviews: Lenny Rachitsky @lennysan shared a definitive guide to analytical thinking interviews, highlighting skills like market context understanding and metric establishment.
AI Industry Developments & News
Meta’s Big AI Impact: Clément Delangue @ClementDelangue argued that Meta’s open-source Llama release transformed the AI landscape and predicted its impact could 100× with addition of @alexandr_wang.
Medical AI Breakthrough: Rowan Cheung @rowancheung reported Microsoft’s new MAI Diagnostic Orchestrator that solved 85.5% of cases vs. 20% by experienced physicians across 304 scenarios.
AI Glossary Update: Lenny Rachitsky @lennysan added Evals to his AI glossary, clarifying new terminology product managers should know.
From YouTube
Create AI Videos with VEO3 on Autopilot! (Make + n8n Templates Inside)
Helena Liu • July 02, 2025
In this tutorial, Helena Liu demonstrates how to automate CGI-level video production using VEO3’s API via FLAI—leveraging ChatGPT in Make.com and n8n to generate scene prompts, schedule 5–15 minute renders, retrieve the MP4 URL, and either log outputs to Google Sheets or post directly to Instagram, Facebook, and YouTube.
Key Takeaways:
- VEO3 costs $250/month or $0.50–$0.75 per second via API (no audio), versus $20–$30 per second for human CGI, with expected price drops to 10–30% in 6–12 months.
- Using one FLAI API key you call VEO3 by specifying only the URL suffix "v3" with application/json headers and a JSON body; each video takes 5–15 minutes to render, requiring Sleep modules (up to 300 seconds) or polling loops.
- Make.com provides native modules for ChatGPT, scheduling, Google Sheets, routers, and social media posting, while n8n requires custom HTTP request nodes for integrations but offers conditional loops and human-in-the-loop approval.
Operators in R | R for Data Analytics Series
Lex Fridman • July 01, 2025
This video demonstrates R’s assignment and core operators—arithmetic (including modulo and exponentiation), comparison (equality and inequality), and logical operators—illustrating operator precedence (PEMDAS) with examples using variables x and y.
Key Takeaways:
- The modulo operator '%%' returns the remainder of a division, so 10 %% 3 yields 1.
- Comparison operators produce boolean values and require '==' for equality (x == y) and '!=' for inequality.
- Logical operators '&' (AND) and '|' (OR) combine boolean expressions—'&' returns TRUE only if both conditions are true, while '|' returns TRUE if at least one is true—and '!' negates boolean values.
Claude Code Hooks is AMAZING: "Text Message Me When AI Agent is Done"
All About AI • July 01, 2025
In this video, All About AI showcases how to customize and extend Claude Code with its new Hooks feature. The host demos stop hooks that send Twilio SMS notifications when tasks complete, post-tool-use hooks that auto-generate Python unit tests, shell-command hooks for auto-staging, committing, and pushing Git changes, and even custom sound alerts.
Key Takeaways:
- A stop hook using a cURL command and Twilio integration sends a text message ("Claude Code has completed the task") each time an AI agent stops, demonstrated with an HTML/CSS ping-pong simulator.
- A post-tool-use hook automatically generates and runs Python unit tests for new code files, shown by creating a prime checker test and validating its behavior.
- Shell-command hooks defined in settings.json can autonomously stage, commit, and push changes to a GitHub repository—and even revert to a specific commit hash—removing manual checkpointing.