Brief IA
›

Google AI: Tulsee Doshi Details the 60-20-20 Framework

AI Tools·Tom Levy·

Google AI: Tulsee Doshi Details the 60-20-20 Framework

Google AI: Tulsee Doshi Details the 60-20-20 Framework
Key Takeaways
1Tulsee Doshi, product lead at Google, oversees Gemini and other multimodal models
2She states that the role of product managers is becoming increasingly important in AI
3She proposes a 60-20-20 framework to balance development choices
Why it matters : Developing cutting-edge AI models requires complex trade-offs, where the expertise of product managers becomes a central asset.

Tulsee Doshi, product lead for cutting-edge models at Google, asserts that product managers are gaining importance in AI. She describes a 60-20-20 framework to guide trade-offs and presents a full-stack approach for models like Gemini and other multimodal systems.

Rising PM Importance and the 60-20-20 Framework, According to Tulsee Doshi

Tulsee Doshi claims that the value of skilled product managers is significantly increasing in AI. She believes that in cutting-edge labs, PMs navigate sometimes competing priorities and needs. To guide these choices, she proposes a 60-20-20 framework aimed at balancing what is already working and what is nearly operational. She also provides advice on how to draft an initial model assessment.

Le brief IA que lisent les pros

Le brief IA que les pros lisent chaque soir

Les 7 actus IA du jour, décryptées en 5 min. Gratuit.

Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.

Choisissez votre rythme

Gratuit · Pas de spam · Désabonnement en 1 clic

A Full-Stack Approach and a Role Beyond Engineering

Tulsee Doshi describes Google's strategy for its cutting-edge models as full-stack. She argues that model development is not solely the domain of engineering. To illustrate this complexity, she compares this work to a potions class taught by Severus Snape.

The Scope Covered: Gemini and Several Multimodal Models

Tulsee Doshi is the product lead for Google's cutting-edge AI models, including Gemini. She also mentions several relevant multimodal models: Nano Banana, Genie, Veo, and Lyria.

Brief IA — L'actualité IA en français

L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.