When agents need to keep learning - Roberta Raileanu, Google DeepMind
From Air Street Press by Nathan Benaich (Air Street Capital)
March 30, 2026 · 6 min
About this episode
Roberta Raileanu discusses her research on AI agents and their ability to continuously learn and acquire new skills.
Roberta Raileanu leads open-ended learning at Google DeepMind and co-authored Toolformer. From RIDE and AMIGo to Llama 3's tool use and MLGym, her research tackles what it takes for AI agents to keep acquiring skills. A RAAIS 2026 speaker profile from Air Street Press.
People in this episode
Host: Nathan Benaich
Guest: Roberta Raileanu
Topics covered
- AI agents
- open-ended learning
- skill acquisition
- machine learning
- research
Keywords
- AI
- learning
- Google DeepMind
- Roberta Raileanu
- Toolformer
- RIDE
- AMIGo
- Llama 3
- MLGym
Mentioned in this episode
Organizations: Google DeepMind
Books & works: Toolformer
More episodes of Air Street Press
- Alta Ares: the Iron Dome for autonomous air defense · June 9, 2026 · 8 min
- Speech-to-text at conversation speed - Angelos Perivolaropoulos, ElevenLabs · May 31, 2026 · 7 min
- From clip-makers to simulators: Odyssey's new world models · May 19, 2026 · 5 min
- What it takes to ship AI under regulation - Nikolay Donets, Revolut · May 17, 2026 · 5 min
- State of AI · May 2026: cyber threshold, China parity, agents in real markets · May 4, 2026 · 28 min
- The next gene editor will be designed: Profluent + Lilly, $2.25B · April 28, 2026 · 6 min
Explore listener stats, chart rankings, contacts and more on the Air Street Press podcast page.