Rate Limited

Ray Fernando, Adam (GosuCoder), and Nathan Snell are back on Rate Limited. Adam has Grok Bot logging into QuickBooks to run payroll for his side business and managing eBay and TCGPlayer accounts, while the crew digs into AI fatigue from too many bot messages, whether sub-agents or flat agents win in practice, and a blunt take on whether the engineering job market is cooked — plus hard-won advice for people just trying to break in. Links Ray Fernando: https://www.youtube.com/@RayFernando1337 Adam (GosuCoder): https://www.youtube.com/@GosuCoder Chapters 00:00 Welcome to Rate Limited 01:10 Grok Bot is running Adam's payroll 02:17 Grok Bot cost, plans, and usage limits 04:34 Plugins, elder care, and the vault 06:23 Bot message fatigue 06:51 Orchestrating bots (meet Grocky) 08:48 Decision fatigue after AI-heavy days 11:40 Context switching vs real productivity gains 15:03 Why big companies struggle with AI 17:45 Betting on a conversational harness 21:56 Sub-agents vs flat agents 24:06 Wave engineering and task decomposition 28:08 Vertical harnesses, skills, and Conway's Law 30:36 Job market despair and generational luck 37:09 Early internet lessons: jobs lost vs created 40:58 Is the engineering job market cooked? 44:10 Mono apps, MCP, and what's next Show notes / key topics • Grok Bot for payroll (QuickBooks), eBay, and TCGPlayer • Plan pricing, usage burn, and coding vs ops workflows • Vault for credentials at runtime • Bot inbox / decision fatigue and orchestrator bots • Token spend vs productivity • Sub-agents vs flat agents; wave engineering • Conway's Law and org-shaped agent designs • Engineering job market anxiety and adapting • Mono apps, MCP, SMS/email agents Tags Rate Limited, Grok Bot, AI agents, sub-agents, flat agents, AI fatigue, decision fatigue, payroll automation, QuickBooks, job market, engineering jobs, AI native, harness engineering, Ray Fernando, GosuCoder, Nathan Snell, podcast, Ep 24

In this episode, the hosts explore the latest advancements in AI, including GPT models, game development in Claude, and the impact of AI on content creation and automation. They discuss practical applications, model performance, and the future of AI-driven workflows.Links:Ray: https://www.youtube.com/@RayFernando1337Adam: https://www.youtube.com/@GosuCoderChapters00:00 Introduction to AI advancements and topics02:16 Exploring multiplayer game development and AI's role03:22 Impressive AI-generated game visuals and physics in Sonnet 5.504:22 The significance of running AI in browsers and low-code environments05:37 AI in world-building and simulation games07:17 Father-son collaboration in game development and AI's educational potential09:18 Impact of Astra and Opus 5.5 on game development and productivity11:17 Using AI for rapid game prototyping and multiplayer features12:24 AI's role in video editing and content automation15:27 The shift from skills-based to model-based AI workflows18:32 Ray's insights on Claude and Opus 5.5 models20:51 AI for video editing and content repurposing23:38 AI for automated content creation and social media clips26:15 The future of AI models and cost considerations27:51 The importance of collaborative AI development and speed28:55 Discussion on AI tools like Grokbot, Muse, and others31:16 Concerns about data privacy and tech accountability33:23 The evolving landscape of AI models and local inference40:45 Ray's experience with local models and hardware setup43:34 Closing thoughts and future outlook on AI innovation

In this episode, Ray Fernando, Adam GosuCoder, and Nathan Snell explore the latest in AI tools like Grokbot, Astra, and Gemini models, sharing insights on how these innovations are transforming productivity, game development, and personal management.Links:Ray: https://www.youtube.com/@RayFernando1337Adam: https://www.youtube.com/@GosuCoderDr. Eggbot: https://x.com/poteto/status/2094967827019243547?lang=en

In this episode, the panel discusses the latest developments in AI watermarking, attribution, and agentic runtimes, exploring their implications for security, privacy, and software development. They share insights on new models, tools, and the future of AI-driven workflows.Links:Ray: https://www.youtube.com/@RayFernando1337Adam: https://www.youtube.com/@GosuCoderkey topicsAI watermarking and regulationAttribution as a new form of AI trackingAgentic runtimes and their role in automationThe latest AI models like GPT-5.6 and Claude 4.6Open source AI community and hardware advancementsAI workflow orchestration and self-managing agentsImplications for security, privacy, and complianceFuture trends in AI development and deploymentTools like Grok bot, Cursor, and Codex for automationThe philosophical and ethical considerations of AI autonomyChapters00:00 - Intro: Welcome to the Rate Limited Podcast00:47 - The Claude Watermarking Controversy06:19 - Is AI Watermarking Just the Next Wave of Attribution?11:31 - IDE Vibe Check: Why Ray is "Cursor Pilled"15:32 - Why Nathan Switched Fully to GPT 5.6 Soul & Codex18:19 - Building a Fully Automated Game Art Pipeline22:23 - The Dream Project: A 100% Agent-Managed Codebase24:18 - Deploying Specialized AI Teammates with Grokbot28:32 - Testing the Grok Build CLI on Large Codebases31:12 - Mind-Blowing Agentic Runtimes (The Atomic Framework)36:55 - How to Lock Down Deterministic Tasks in a Non-Deterministic AI World46:39 - Why Traditional AI Evals Don't Actually Work49:31 - Local Hardware: Running DeepSeek V4 Flash on DGX Spark53:02 - Outro & Community Shoutouts

We explore the future of AI, its impact on engineering, and how non-technical users are leveraging AI in everyday life. We discuss models like GPT-5.6, the importance of system thinking, and opportunities for solo entrepreneurs.topicsAI model evolution and comparison (GPT-5.6, Claude, Grok 4.5)The gap between technical and non-technical AI usersOpportunities for solo entrepreneurs in AISystem thinking and engineering best practices in AI developmentRisks and safety considerations in AI code generationVerticalization of AI applications across industriesThe future of skills and AI in the workplaceAI fatigue and user experience designChapters00:00 Introduction and overview of AI's impact on engineers01:59 The gap between technical and non-technical AI users03:56 AI in marketing and outside engineering05:06 The importance of system thinking in AI development08:10 Verticalization of AI applications across industries

In this episode, industry experts discuss the latest in AI advancements, including the Fable Five jailbreak, SpaceX's acquisition of Cursor, and evolving software development practices. They explore how AI is transforming workflows, the importance of verification, and the future of AI-powered tools for both technical and non-technical users.Links:Ray: https://www.youtube.com/@RayFernando1337Adam: https://www.youtube.com/@GosuCoderkey topicsAI advancements and implicationsFable Five jailbreak and security concernsSpaceX's acquisition of Cursor and product announcementsThe shift from traditional PRDs to prototypes in product managementThe role of verification and success criteria in AI developmentThe concept of skills, harnesses, and agents in AI workflowsThe impact of rapid software updates and UI changes on user experienceThe importance of focused versus generic AI harnessesThe future of AI tools for both technical and non-technical usersChapters00:00 Introduction to the Rate Limited Podcast01:00 Exploring Fable Five and Its Implications05:02 The Future of Product Requirements Documents (PRDs)09:55 SpaceX's Acquisition of Cursor and Its Impact14:44 The Evolution of Skills and Harnesses in AI26:24 The Ease of Use in Technology29:21 Understanding the Software Factory Concept32:32 The Evolution of Startups and Software Development37:40 The Impact of Rapid Changes in Software Interfaces40:33 The Challenge of Automation vs. Human Efficiency44:49 The Excitement of Innovation and Future Possibilities

In this episode, we explore the latest developments in AI, including cost dynamics, looping strategies, and the impressive capabilities of models like Fable 5. We  discuss operational challenges, the future of AI-driven software factories, and how to navigate this rapidly evolving landscape.Links:Ray: https://www.youtube.com/@RayFernando1337Adam: https://www.youtube.com/@GosuCoderEric: https://www.youtube.com/@pvncherChapters00:00 Introduction to Rate Limited Podcast01:35 The Cost of AI: A Nuanced Discussion07:40 Token Maxing and ROI in AI Usage16:04 The Evolution of AI: From Prompts to Loops19:09 The Value of Proactive AI Engagement22:48 Exploring Fable Five: A New Frontier in AI26:51 Diverse Perspectives on AI Models30:54 Future of Work in an AI-Driven World

Google I.O. brought major updates, but the developer community is furious. Is Gemini 3.5 Flash a massive step backward for production apps? In Episode 16 of Rate Limited, Ray, Eric, and Adam break down the massive backlash surrounding Google’s new pricing model and token consumption, why the Gemini CLI is getting aggressively sunset, and whether Cursor’s new Composer 2.5 is the absolute pinnacle of AI coding workhorses. We also react to Andrej Karpathy’s massive jump to Anthropic, detail a wild success story of modding Zelda into VR using Codex's Goal Mode, and discuss how to transition your mindset into the era of probabilistic, agentic engineering.If you are an engineer or builder navigating the frontier of AI, hit that subscribe button for deep, unfiltered technical breakdowns twice a month.Links:Ray: https://www.youtube.com/@RayFernando1337Eric: https://www.youtube.com/@pvncherAdam: https://www.youtube.com/@GosuCoder00:00 - Google I.O., OpenAI, & Composer 2.500:45 - The Truth About Gemini 3.5 Flash & Token Guzzling02:08 - Is Gemini Rebranding What "Flash" Means?03:45 - Prompt Performance & Instruction Following in Anti-Gravity05:47 - Google's Profit Margins and Capacity Constraints06:15 - RIP Gemini CLI: Google Consolidating Compute07:38 - The Hidden Cost of Small Model Reasoning09:38 - Google's Distillation Strategy & Compute Allocation11:41 - Knowledge Cutoffs & Eval Degradation in Gemini 3.513:40 - Anti-Gravity 2.0 Drama: Shifting Away From Consumers16:40 - Composer 2.5 First Impressions: A Pure Coding Workhorse19:15 - Building Parallel QA Agents with Browser Use21:32 - How Cursor Pulls Off Trillion-Parameter Speeds23:18 - Is Speed the Moat? Continuous Improvement RL Loops25:25 - The Messiness of Real-World Context vs. Static Evals27:13 - Tool Search vs. Preloading Context Bloat28:14 - Minimal Context Setups vs. Skill Maximalism32:00 - The Lack of Portability Between AI Harnesses34:25 - GPT-5.5 Low/Medium vs. Anthropic 4.7 Latency38:10 - Tool Tinkering vs. Grabbing Off-The-Shelf Tech39:11 - The Team's Current Model Stack (Vibe Check)42:11 - Codex App: Mind-Blowing Background Computer Use44:39 - Goal Mode vs. High-Level Orchestration Loops48:23 - Success Story: Modding Zelda into VR Using Goal Mode50:33 - Andrej Karpathy Joins Anthropic: Why Now?52:30 - The Value of Co-location & The Bay Area Sparkle55:00 - Impact Over Upside: Technologists Shaping Green Spaces56:09 - The Mindset Shift: Transitioning to Agentic Engineering58:30 - Auditing Your Threads & Finding Inefficiencies01:00:27 - AI Automation as "Mana" for Your Life

This episode explores the latest developments in AI models like GPT 5.5, their impact on workflows, trust, and the future of agent development. Featuring insights from industry experts, it covers model performance, rate limits, AI in enterprise, and practical tips for building effective AI agents.Links:Ray: https://www.youtube.com/@RayFernando1337Eric: https://www.youtube.com/@pvncherAdam: https://www.youtube.com/@GosuCoderChapters00:00 Introduction to the Rate Limited Podcast00:45 Exploring GPT 5.5: Features and User Experiences04:28 Switching from Anthropic to GPT 5.5: User Insights07:17 Trust and Performance: Comparing Models10:45 Improving Code Quality with GPT 5.517:48 Utilizing Goal Mode for Long-Term Tasks21:26 Anthropic and SpaceX: A New Partnership25:32 The Future of AI Automation on Mac30:51 The Impact of AI on Layoffs36:35 Navigating the AI Landscape41:45 The Role of Engineers in AI Development49:23 Creating Engaging AI Agents55:03 The Future of Agent Development

This episode explores the latest AI model updates, including Opus 4.7, Kimi 2.6, and the evolving landscape of AI in design, coding, and enterprise applications. The hosts discuss model performance, industry trends, and strategic moves like SpaceX's partnership with Cursor, providing insights into the future of AI development.Links:Ray: https://www.youtube.com/@RayFernando1337Eric: https://www.youtube.com/@pvncherAdam: https://www.youtube.com/@GosuCoderChapters00:00 Introduction to the Podcast and Recent Developments02:49 Exploring Opus 4.7: Impressions and Comparisons05:58 The Art of Prompting: Strategies for Effective Use08:57 Design Innovations: Claude Design and Its Impact12:03 The Future of Design: AI's Role and Implications15:00 Kimi 2.6: Performance and Comparisons with Other Models22:17 Exploring Composer's Efficiency25:19 Kimi 2.6 Performance Insights30:44 The Impact of Tokenization on AI Models33:15 Cursor and SpaceX: A Strategic Partnership40:20 The Future of AI Coding Companies46:24 The End of RooCode: Reflections on Community46:32 The Evolution of AI Coding Practices50:15 Navigating AI Engineering Workflows56:24 Orchestrating AI Agents for Efficiency01:02:00 The Future of AI in Development