Beyond the Chatbot: A Blueprint for Trustable AI
Talk at Google DeepMind on building trustable AI systems beyond simple chatbots. **Key points:** - Moving from conversational AI to agentic AI architecture - Safety frameworks for …
From Whiteboard to Users: Making Research Accessible
Talk at Google I/O GDE Summit 2025 on making academic research accessible to practitioners. **Key points:** - Translating PhD research into real-world products - Lessons from …
Deep Dive into the Google Agent Development Kit (ADK): Features and Code Examples
In our previous overview, we introduced the Google Agent Development Kit (ADK) as a powerful Python framework for building sophisticated AI agents. Now, let's dive deeper into some …
LLM, Reasoning and Agentic Gemma
Talk at Google Korea on LLM reasoning and agentic Gemma. **Key points:** - Building agentic AI systems with Google Gemma models - Reasoning and planning capabilities of …
LLMinABox: On Device Personalized Diary & Concierge using your voice and Gemma
Talk at Google Mountain View on LLMinABox, an on-device AI diary and concierge. **Key points:** - On-device LLM with Google Gemma for privacy - Voice-driven interaction for …
On Device Generative AI: Building your own Dall-E in the browser, welcome WebGPU
Talk at Google San Jose Meetup on building on-device generative AI in the browser. **Key points:** - Running image generation models in the browser with WebGPU - TensorFlow.js for …
DeepSpeech: A Journey to <10% Word Error Rate
Talk on DeepSpeech and the journey to achieving less than 10% word error rate in automatic speech recognition. **Key points:** - Mozilla DeepSpeech architecture and training - …