Francis Clase

Alignment Science

AI Safety & Alignment

Ensuring AI systems pursue intended goals and values through rigorous scientific methodology.

Reward Models

AI Safety & Alignment

Developing systems that learn what humans value through preference modeling and feedback.

Alignment Finetuning

AI Safety & Alignment

Refining AI models to better align with human preferences and safety constraints.

Model Welfare

AI Safety & Alignment

Examining ethical considerations regarding the treatment and experiences of AI systems.

Honesty in AI Systems

AI Safety & Alignment

Research into making AI systems truthful, calibrated, and transparent in their outputs.

Pre-training

Model Training & Optimization

Training large-scale foundation models on vast datasets before task-specific fine-tuning.

Pretraining Scaling

Model Training & Optimization

Understanding how model capabilities improve with increased compute, data, and parameters.

RL Engineering

Model Training & Optimization

Building reinforcement learning systems that learn from interaction and feedback.

Production Post-Training

Model Training & Optimization

Optimizing deployed models through continued learning and adaptation in production.

Performance Engineering

Optimizing AI systems for speed, efficiency, and resource utilization at scale.

GPU Performance Optimization

Performance Engineering

Maximizing computational efficiency on graphics processing units for AI workloads.

Model Performance & Quality

Performance Engineering

Measuring and improving both computational efficiency and output quality of AI systems.

Interpretability

Research & Analysis

Understanding the internal workings and decision-making processes of AI systems.

Multimodal AI

Research & Analysis

Developing AI systems that understand and generate multiple types of data (text, images, audio).

Token Research

Research & Analysis

Studying how AI systems represent and process discrete units of information.

Machine Learning Horizons

Research & Analysis

Exploring emerging frontiers and future directions in machine learning research.

Biology & Life Sciences AI

Research & Analysis

Applying AI to biological research, drug discovery, and understanding living systems.

ML Acceleration

ML Infrastructure & Systems

Developing specialized hardware and software to speed up machine learning computations.

Inference Systems

ML Infrastructure & Systems

Building efficient systems for deploying and running AI models in production.

ML Networking

ML Infrastructure & Systems

Optimizing network infrastructure for distributed machine learning training and inference.

TPU Kernel Engineering

ML Infrastructure & Systems

Writing low-level code for Tensor Processing Units to maximize AI computation efficiency.

Accelerator Build Infrastructure

ML Infrastructure & Systems

Creating tools and systems for developing and testing AI accelerator hardware.

Data Operations

Operations & Discovery

Managing data pipelines, quality, and infrastructure for machine learning systems.

Research Tools

Operations & Discovery

Building software infrastructure to accelerate AI research and experimentation.

Discovery & Innovation

Operations & Discovery

Exploring new research directions and identifying breakthrough opportunities in AI.