Lukas Biewald co-founded Weights & Biases in 2017 after an unpaid internship at OpenAI. The company started behind a karate studio in San Francisco to help engineers track machine learning models. Biewald previously led CrowdFlower, which sold for $300 million in March 2019. He now builds tools for AI developers.

Synopsis

After reportedly leaving the CEO role, Lukas Biewald felt that he needed to strengthen his technical knowledge as deep learning advanced. He then spent time building robots, experimenting with TensorFlow in a Raspberry Pi and studying neural networks. He taught introductory deep-learning classes to thousands of engineers, which helped him understand the problems developers faced when building AI models.

Lukas Biewald co-founded CrowdFlower in 2007 to solve a major problem in machine learning. The aim was to obtain enough accurately labelled data to train AI systems. The platform combines human input with automation to label data used in machine-learning applications.

The company later changed its name to Figure Eight and worked with major businesses, including eBay and Google. Lukas Biewald remained involved with the company for more than a decade but eventually stepped down as CEO as he began to focus on the technical side of AI again.

Figure Eight was acquired by Appen in March 2019 for around $300 million, according to Insight Partners.

Why did he return to learning artificial intelligence?

After reportedly leaving the CEO role, Lukas Biewald felt that he needed to strengthen his technical knowledge as deep learning advanced. He then spent time building robots, experimenting with TensorFlow in a Raspberry Pi and studying neural networks.

He taught introductory deep-learning classes to thousands of engineers, which helped him understand the problems developers faced when building AI models.

Live Events

He also arranged an unpaid internship at OpenAI, where he gained first-hand exposure to the challenges of developing machine-learning systems. This helped him identify a gap in the market. He saw that AI developers lacked effective tools to track experiments, compare models and reproduce results.

The company stared behind a karate studio

In 2017, Lukas Biewald co-founder Weights & Biases with Chris Van Pelt, his longtime CrowdFlower co-founder, and Shawn Lewis, a former Google engineer. The company began behind a karate studio in San Francisco, according to its official company history.

Its original goal was to build better software tools for machine-learning researchers and engineers.

The founders developed an experiment-tracking tool that allows AI developers to monitor their models, compare training runs and understand how changes affected performance. The company's early work is closely connected to OpenAI. Biewald initially built the tool to address problems faced by people working on machine-learning projects there.

It gained attention as AI development accelerated and more teams needed reliable ways to manage increasingly complex models.

How a small tool became a big AI developer platform

Weights & Biases expanded beyond its initial OpenAI user base as other companies began adopting deep learning. Its software helps teams track experiments, compare model versions, identify problems and make AI development more reproducible.

By 2023, the company had reached a valuation of $1.25 billion following a funding round, reported TechCrunch.

On March 4, 2025, CoreWeave announced an agreement to acquire Weights & Biases. It provides cloud computing infrastructure designed for AI workloads. The deal's reported value was approximately $1.7 billion, according to The Information and TechCrunch.

KYB and KYC: Quick commerce, the Damani premium and what DMart must deliver – 3rd and concluding part

The tough business behind the walls

KYC and KYB: Inside Damani’s DMart, the retail chain for the middle class. And stock for the long term? – Part 2 of 3

Will it be a regulatory nightmare if an exchange lists on its own platform?

KYC and KYB: Inside Damani’s Avenue Supermarts that refuses to charge more – Part 1 of 3

Will oil turn into a lingering problem? Is the world economy, India, or the market prepared?