Episode 269 - The "Big Three" AI Models and Training Evolution

<p>In Part 1 of a two-part series, guest host Sam Witteveen joins Allen to catch up and dive deep into the rapidly evolving world of AI models. Sam shares his fascinating journey from being a successful pop songwriter to becoming a Machine Learning Google Developer Expert (GDE) and running the massive Machine Learning Singapore meetup.</p><p><br></p><p>The conversation shifts to the latest AI developments, exploring the &quot;Big Three&quot; model builders—Anthropic, OpenAI, and Google. Sam and Allen discuss the frenetic pace of new model releases, changes to the Gemini 3 API, and how developers navigate the trade-offs between intelligence, latency, and cost.</p><p><br></p><p>Finally, they pull back the curtain on how these models are actually trained today. Discover why models are no longer trying to be &quot;fact machines&quot; and how post-training breakthroughs, code execution sandboxes, and Reinforcement Learning (RL) environments are dramatically improving AI capabilities. Stay tuned for the end of the episode, where they hint at what&#39;s coming in Part 2!</p><p><br></p><p>Timestamps:</p><p>[00:00:00] Introduction and catching up</p><p>[00:01:33] Sam&#39;s fascinating journey from pop music to machine learning</p><p>[00:05:23] Running the massive Machine Learning Singapore meetup</p><p>[00:07:42] Stumbling into YouTube and teaching AI with Google Colab</p><p>[00:12:38] Analyzing the &quot;Big Three&quot; AI models and rapid release cycles</p><p>[00:17:52] Gemini 3 API updates, Flash models, and thinking levels</p><p>[00:22:00] Tool use, knowledge cutoffs, and why LLMs aren&#39;t fact machines</p><p>[00:26:00] How post-training and code sandboxes revolutionized AI</p><p>[00:32:00] Scaling Reinforcement Learning (RL) environments for design</p><p>[00:34:04] Structured outputs and the return to predictable rules</p><p>[00:36:43] Tune in next time for more! And where to find Sam online</p><p><br></p><p>Hashtags:</p><p>#TwoVoiceDevs #AI #MachineLearning #DeepLearning #LLM #GoogleGemini #Gemini #OpenAI #ChatGPT #Anthropic #Claude #ReinforcementLearning #RAG #Developers #SamWitteveen</p>