jeff-dean

What problem does it solve?

This skill provides a principled, engineering-focused frame inspired by Jeff Dean to help teams design and optimize large-scale AI systems, balancing frontier capabilities with practical latency, energy, and deployment constraints.

Core Features & Use Cases

• Hardware-Algorithm Co-design mindset to align model architectures with accelerator constraints.
• Latency-first design and Model Distillation to enable scalable, cost-effective deployment across orgs.
• Promote Massively Multi-task models and Unified multimodal approaches to reduce siloed compute.
• Guide system design decisions with a 5-10x scaling horizon to avoid 100x premature scaling.
• Provide decision guidance for transitioning from specialized to generalized models in production.

Quick Start

Initiate by identifying the top bottlenecks in your current ML system using Back-of-the-Envelope thinking, then draft a 5-10x scaling plan that preserves latency and energy budgets.