AI experts by specialty

Research & evaluation experts

Find AI researchers and evaluation specialists working on machine learning, model safety, and reproducible experiments.

Showing 1–12 of 127 experts in Research & evaluation · Featured priority · All fields ↗
Page 1 of 11

Andrej Karpathy

Eureka Labs / ex-Tesla, OpenAI

Andrej Karpathy is an AI researcher and educator who founded Eureka Labs, an AI-native education company. Previously, Andrej served as the Director of AI at Tesla, leading the Autopilot computer vision and neural network architecture teams, and was a founding research scientist at OpenAI. Widely revered as one of the world's premier deep learning educators, he created nanoGPT and the landmark 'Neural Networks: Zero to Hero' lecture series.

Creative AI & educationResearch & evaluation
Sources checked5 artifacts

Abeba Birhane

Mozilla / Trinity College Dublin

Senior Fellow in Trustworthy AI at the Mozilla Foundation and Assistant Professor at Trinity College Dublin. Named by TIME Magazine as one of the 100 most influential people in AI. Cognitive scientist specializing in relational ethics and critical auditing of large-scale multimodal training datasets (such as LAION-5B), uncovering systematic demographic disparities, exploitative content, and copyright vulnerability.

AI Safety, Evals & Red TeamingAI Governance & Compliance
Sources checked2 artifacts

Co-founder and CEO of Cohere, enterprise AI platform for generative LLMs, fine-tuning, and semantic retrieval. Co-authored the landmark 'Attention Is All You Need' paper in 2017 while at Google Brain, which introduced the Transformer architecture that underpins all modern generative AI models. Completed Ph.D. at University of Oxford.

Engineering and InfrastructureResearch and Evaluation
Sources checked2 artifacts

Research Scientist at OpenAI and primary architect of modern generative AI. Lead author of the original GPT (Improving Language Understanding), GPT-2, CLIP, and Whisper, establishing the scaling paradigms for unsupervised pre-training and contrastive vision-language representation.

Research and EvaluationMultimodal & Real-Time Voice AI
Sources checked2 artifacts

Alex Ratner

Snorkel AI / University of Washington

Co-founder and CEO of Snorkel AI, pioneering programmatic data labeling, synthetic curation, and alignment for enterprise foundation models. Assistant Professor of Computer Science at the University of Washington. Completed his PhD at Stanford University in the Stanford DAWN lab, where he created Snorkel to replace manual hand-labeling with programmatic weak supervision, generative labeling functions, and statistical label aggregation.

AI Safety, Evals & Red TeamingEnterprise AI
Sources checked2 artifacts

Founder and CEO of Scale AI, the data infrastructure bedrock supporting frontier model developers including OpenAI, Meta, Google, Microsoft, and the US Department of Defense. Started Scale AI at age 19 after studying at MIT. Creator of the SEAL Leaderboards (Safety, Evaluations, and Alignment Lab), establishing rigorous third-party private benchmark methodologies for coding, reasoning, and adversarial red teaming.

AI Data AnalysisAI Safety, Evals & Red Teaming
Sources checked2 artifacts

Ali Ghodsi

Databricks

Co-founder and CEO of Databricks, the multi-billion-dollar enterprise data and AI company. Adjunct Professor at UC Berkeley EECS. Co-creator of Apache Spark, Delta Lake, MLflow, and the Data Intelligence Platform. Spearheaded Databricks' pivotal acquisitions of MosaicML and Lilac AI, pioneering compound AI systems and enterprise foundation model training that combine proprietary tabular enterprise data with generative LLMs.

Enterprise AIMachine Learning
Sources checked2 artifacts

Founder and CEO of Vectara, co-founder and former CTO of Cloudera (NYSE: CLDR), and former VP of Developer Relations at Google Cloud. PhD in Electrical Engineering from Stanford University. Pioneer of distributed big data systems and modern Agentic Retrieval-Augmented Generation (RAG). Created the Vectara Hughes Hallucination Evaluation Model (HHEM), the industry-benchmark leaderboard measuring factual consistency across foundational LLMs.

Agentic RAG & Knowledge EngineersEnterprise AI
Sources checked2 artifacts

Andrew Trask

OpenMined / Google DeepMind / University of Oxford

Founder and executive leader of OpenMined, Senior Research Scientist at Google DeepMind, and PhD in Computer Science from the University of Oxford. Pioneer of privacy-preserving machine learning, federated learning, differential privacy, and encrypted computation. Author of the acclaimed book 'Grokking Deep Learning', training tens of thousands of software developers in neural network fundamentals from scratch.

AI Safety, Evals & Red TeamingAI Governance & Compliance
Sources checked2 artifacts

Anima Anandkumar

California Institute of Technology (Caltech)

Bren Professor of Computing and Mathematical Sciences at Caltech and former Senior Director of AI Research at NVIDIA. ACM Fellow and IEEE Fellow. World-renowned authority on tensor algebra, non-convex optimization, and AI for science. Pioneer of Neural Operators—specifically Fourier Neural Operators (FNO)—which solve complex partial differential equations (PDEs) for weather prediction, aerodynamics, and carbon sequestration up to 100,000x faster than traditional numerical solvers.

AI ScienceMachine Learning
Sources checked2 artifacts

Aravind Srinivas

Perplexity AI

Co-founder and CEO of Perplexity AI. Former research scientist at OpenAI focusing on language models and contrastive learning, and research intern at DeepMind and Google Brain. Completed Ph.D. in Computer Science at UC Berkeley under Pieter Abbeel, pioneering unsupervised visual representation and sub-quadratic attention systems.

Agents and AutomationEngineering and Infrastructure
Sources checked2 artifacts

Arthur Mensch

Mistral AI

Co-founder and CEO of Mistral AI. Previously a research scientist at DeepMind, working on Retro, Chinchilla, and Flametree architectures. Led Mistral to develop open-weights models that match and exceed proprietary frontier models with efficient grouped-query attention and sparse mixture-of-experts architectures.

Engineering and InfrastructureResearch and Evaluation
Sources checked2 artifacts
Showing 1–12 of 127 experts

Guides to evaluating AI expertise