Rethinking Pre-Training for Agentic AI with Aakanksha Chowdhery

Episode 759

0:00-52:54

Today, we're joined by Aakanksha Chowdhery, member of technical staff at Reflection, to explore the fundamental shifts required to build true agentic AI. While the industry has largely focused on post-training techniques to improve reasoning, Aakanksha draws on her experience leading pre-training efforts for Google’s PaLM and early Gemini models to argue that pre-training itself must be rethought to move beyond static benchmarks. We explore the limitations of next-token prediction for multi-step workflows and examine how attention mechanisms, loss objectives, and training data must evolve to support long-form reasoning and planning. Aakanksha shares insights on the difference between context retrieval and actual reasoning, the importance of "trajectory" training data, and why scaling remains essential for discovering emergent agentic capabilities like error recovery and dynamic tool learning.

Publisher

The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)

Link

twimlai.com/podcast/twimlai/rethinking-pretraining-for-agentic-ai/

Published

December 17, 2025

Duration

53 min

Format

Podcast

Category

Issues · Tech News

Source

RSS feed

Host

Sam Charrington

TT

The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)

@twimlai

#rethinkingpre #agenticai #aakankshachowdhery

The Nile