In this episode, Guy and Simon dive deep into February’s wave of AI model releases, focusing heavily on the showdown between OpenAI’s GPT-4.5 and Anthropic’s Sonnet 3.7. They discuss GPT-4.5's move toward higher emotional intelligence (EQ), how it compares with Claude, and whether it’s actually meeting developers' expectations—or falling short.

Anthropic’s introduction of dynamic reasoning with Sonnet 3.7 gets special attention as Guy and Simon explore what this new feature means practically for developers, referencing Bolt.new’s recent implementation as an example. They also weigh the strategic choices these AI labs face—when to build tools themselves and when to act purely as platforms—highlighting the delicate balance companies like Anthropic and ElevenLabs must strike.

The conversation also covers practical developer workflows, featuring insights from recent interviews with Farhath, who leverages multiple AI dev tools simultaneously, and Mati from ElevenLabs, who shares how his company navigates scaling and research while remaining nimble.

Finally, Guy and Simon announce AI Native Dev Con: Spring 2025, and hint at exciting upcoming episodes, rounding out a packed discussion that goes straight to the heart of what matters for AI-native developers today.

Watch the episode on YouTube: https://youtu.be/kjAs_isf-Vs

Join the AI Native Dev Community on Discord: https://tessl.co/4ghikjh

Ask us questions: podcast@tessl.io