About us

At Weights & Biases our mission is to build the best tools for machine learning. Our co-founders built Figure Eight, and our tools are being used by cutting-edge machine learning teams including OpenAI, NVIDIA, and Co:here. We’re looking for passionate, driven people who love to learn and collaborate in an inclusive work environment.

Our start

Before Weights & Biases, founders Lukas Biewald and Chris Van Pelt had already ridden the start-up train together building Figure Eight. When pondering about their next project, Weights & Biases came to fruition during Lukas’ internship at OpenAI.

While working with some of the world’s most renowned machine learning researchers, Lukas realized the tools they were using were outdated and inefficient. To build better models, they needed better tools.

Launching a new company is hard, so to round out the team, Lukas and Chris pulled in Shawn Lewis — a long-time friend who also had a founder’s experience. Together, they used their combined knowledge, skills, and experiences to create tools that could support machine learning practitioners and move forward the state of the art of machine learning.

Our values

Honesty

To make smart decisions, we need to be honest with each other and with ourselves. Honesty allows us to surface problems so we can tackle them together as a team, instead of avoiding them.

Curiosity

Machine learning is advancing fast. Best practices can easily become obsolete overnight. The key to thriving (not just surviving) in this industry is to never stop learning.

Gumption

In a world of intense competition, having the confidence to move forward despite uncertainty is essential. Taking risks can be uncomfortable but staying complacent is rarely the better option.

Grit

As much as talent counts, effort counts twice. Having that ability to roll up your sleeves converts setbacks into stepping stones. In the long run, having grit is a core part of success.

Want to learn more about our purpose and values at Weights & Biases? Take a look at this article written by our CEO, Lukas.

Our leadership

Lukas Biewald

CEO and Cofounder

Chris Van Pelt

CISO and Cofounder

Shawn Lewis

CTO and Cofounder

Yan-David Erlich

COO

Cameron Kinloch

CFO

Phil Gurbacki

VP of Product

Lavanya Shukla

VP of Engineering

Adrian Swanberg

VP of Engineering

Our team

Made up of 180 (and counting) talented, kind, and mission-driven individuals, our team is distributed all over the world. If you’re the owner of an inquisitive mind, a kind soul, sees challenges as opportunities, and are committed to going the extra mile — let’s talk.

Our opportunities

If you’d like to join our growing team at W&B, we’d love to hear from you. You’ll find our open roles below.

Admin

Admin - Other

Commercial Counsel (Remote) London

Customer Success

Success ML Engineer - Enterprise

Machine Learning Engineer, Customer Success (Remote) San Francisco, California

Marketing

Head of Demand Generation (Remote) San Francisco, California

R&D

Sales Administration and Business Development

Revenue Operations

Business Systems Analyst (Remote) San Francisco, California

Our investors

We’re a Series C startup with over 150 million dollars in funding. We’re lucky to have the support of some of the most forward-thinking investors in the world, including Coatue, Felicis Venture, BOND, Insight Partners, Bloomberg Beta, NVIDIA, and more.

Our offices

Germany

Weights & Biases
Stresemannstraße 123
Berlin, BE 10963

U.S. Headquarters

Weights & Biases
1479 Folsom Street
San Francisco, CA 94103

Japan

Weights & Biases
Tokyo Square Garden
14F, 3 Chome-1-1 Kyobashi, Chuo
City, Tokyo 104-0031

What is your current MLOps Maturity?

Assess how far your organization has to go to reach the elite level of MLOps maturity.

Try Weights & Biases

Build vs. Buy

Weighing the value of in-house development against a platform trusted by the world’s leading ML teams

In-House Solution

Weights & Biases

ROI of ML projects

The inability to handle a large number of experiments could drastically affect productivity and result in developing less performant models
Real-world ML systems commonly incur massive ongoing costs due to the maintenance problems of traditional code plus an additional set of ML-specific issues

Capable of running thousands of experiments iteratively and collaboratively to build successful models fast and deploy them into production
Automatically track and optimize GPU utilization alongside model performance metrics which can generate 30-50% utilization gains

Just a few of the teams on W&B Saas Cloud

Explore the Weights & Biases platform

The Weights & Biases end-to-end AI developer platform

Weave

Traces

Debug agents and AI applications

Evaluations

Rigorous evaluations of agentic AI systems

Playground

Explore prompts
and models

Agents

Observability tools for agentic systems

Guardrails

Block prompt attacks and harmful outputs

Monitors

Continuously improve in prod

Models

Experiments

Track and visualize your ML experiments

Sweeps

Optimize your hyperparameters

Tables

Visualize and explore your ML data

Core

Inference

Explore hosted, open-source LLMs

Registry

Publish and share your AI models and datasets

Artifacts

Version and manage your AI pipelines

Reports

Document and share your AI insights

SDK

Log AI experiments and artifacts at scale

Automations

Trigger workflows automatically

The Weights & Biases platform helps you streamline your workflow from end to end

Models

Experiments

Track and visualize your ML experiments

Sweeps

Optimize your hyperparameters

Registry

Publish and share your ML models and datasets

Automations

Trigger workflows automatically

Weave

Traces

Explore and
debug LLMs

Evaluations

Rigorous evaluations of GenAI applications

Core

Artifacts

Version and manage your ML pipelines

Tables

Visualize and explore your ML data

Reports

Document and share your ML insights

SDK

Log ML experiments and artifacts at scale

The best machine learning teams in the world use Weights & Biases

No matter the size of your ML team, Weights & Biases enables deep collaboration, logging everything you need to build better models faster. Learn why teams like OpenAI, Stability, NVIDIA, and more trust W&B.

Trusted by some of the most innovative teams in machine learning

“Saving everything in your model pipelines is essential for serious machine learning: debugging, provenance, reproducibility. W&B is a great tool for getting this done.”

“We were drawn to W&B because we realized our existing approach just didn’t work with a remote team. W&B is a much better home for our experimentation results. Plus it’s super easy to use. ”

“W&B lets us examine all of our candidate models at once. We can identify which model produces state of the art results on our robust test suite. This is vital for understanding which model will work best for each customer.”

A single shared source of truth

Unify everything from models and pipelines to experiments and datasets in a single system of record for your ML team. W&B lets you log and track what’s important for your unique use case so your team can be confident about every step of your model development pipeline.

Integrates with the tools and frameworks you’re using already

Weights & Biases plays well with others. We support all major ML frameworks, training and inference environments, and workflow orchestration tools. No matter what you’re using, we’re confident you can use it with W&B.

Production-grade MLOps at scale

Trusted by 400,000+ ML practitioners

Rapid deployment across teams

Track, tune, analyze, and visualize your models in a platform built for collaboration. Weights & Biases lets you run experiments together and derive insights in real-time, automating the training and evaluation processes that slow down large teams so you get to production faster.

Easily access the compute you need, with none of the complexity

There shouldn’t be friction in modern ML teams and workflows; with W&B Launch, there’s won’t be! Easily connect to the high-speed, specialized hardware that you’ve made investments in, and that you need to scale up and out your ML activities. Accelerate training, spin up model evaluation suites, load models for inference, and gain oversight and governance over your big ticket ML infrastructure with W&B Launch.

Tap into a rich partner network

We’ve formed trusted partnerships that accelerate the impact of your ML efforts. Our joint integration investments with cloud providers, AI hardware, ISVs, and ML frameworks enrich our customers’ experience and extend the ML value stack.

VIEW OUR GROWING PARTNER NETWORK‍

MLOps Whitepaper

Read how building the right technical stack for your machine learning team supports core business efforts and safeguards IP

Academic research teams

Use W&B to coordinate projects remotely, like Google Drive for machine learning.

Managed W&B Private Cloud

Need HIPAA compliance, or have data that can’t leave your system? Install W&B locally, on your own servers.

Why use Weights & Biases?

1

Automatic Tracking

Tired of pasting results into a spreadsheet? Track models automatically.

2

Model Debugging

Debug model issues quickly with logs, charts, and tables of tracked results.

3

Collaboration

Quickly share findings and discuss model results with your team.

4

Reproducibility

Use W&B’s reliable system of record to make all models reproducible.

Lineage & Tracking

Storage and Artifacts are free up to 100GB, then usage is billed monthly.
You have complete control of what is logged in W&B. You can curate and delete data from the system at any time.

Storage

The hosting costs of files saved to W&B servers

100 GB included free
$0.08 per GB up to 10 TB
$0.06 per GB up to 100 TB
$0.05 per GB up to 1000 TB

Artifacts

Files explicitly tracked with artifacts for reproducibility

100 GB included free
$0.05 per GB up to 10 TB
$0.03 per GB up to 100 TB
$0.02 per GB up to 1000 TB

Custom plans available.

Key Values of
Weights & Biases

How our lightweight, interoperable tools empower your ML team

Reliable, automatic experiment tracking

Develop better models faster

Enable faster model development

Use the single, central system of record to save all the relevant metadata for your models automatically, so you can focus on model training. Spend more time in a flow state.

Modular, interoperable solutions

Pick the tools that solve your problems best. With W&B, you’re not locked in to an end-to-end platform. We pride ourselves on making our tools play well with other systems.

Stay focused on the hard problems

Spend more time in a flow state, and less time doing tedious tracking. With W&B, you can focus on the hard machine learning problems, and let us take care of tracking all the details to make the models reproducible.

Key insights from team management

How are the latest experiments doing, compared to previous model versions?
Is this model running out of GPU memory? What are the system bottlenecks?
How does this hyperparameter affect accuracy across different classes?
What do sample predictions look like from this model?

Persistent knowledge base

Capture valuable insights centrally

A lab notebook for every ML experiment

Every time you run a new experiment, W&B captures the changes you made and gives you a place to jot down notes. Quickly compare your latest results with previous baselines.

Reproduce any experiment

Automatically track the exact version of the code, hyperparameters, and even the dataset your model trained on. Trace back exactly where your resulting models came from.

Transparent, understandable results

Annotate findings inside the central W&B system, so your models can speak for themselves. Use interactive notes, comments, and reports to clearly explain your research.

Key insights from a persistent knowledge base

Where are all the model files stored for this project?
What have we tried already, and what avenues should we explore next for improving this model?
What are the key findings from this research?

Machine learning team management

Move quickly on big projects, and make handoffs seamless

Easily track anywhere, on any infrastructure

Use the single, central system of record to save all the relevant metadata for your models automatically, so you can focus on model training. Spend more time in a flow state.Train on whatever compute is available – AWS, GCP, Azure, or the GPU box in the office. All results are organized in a central place.

Modular, interoperable solutions

Pick the tools that solve your problems best. With W&B, you’re not locked in to an end-to-end platform. We pride ourselves on making our tools play well with other systems.

Stay focused on the hard problems

Key insights from team management

How should new team members get started contributing to this ML project?
When someone leaves the team, how is their research saved and communicated?
How are all our projects doing, across the organization? Are any particular projects roadblocked or stuck? Are certain projects performing better than expected?

Production model lineage

Version and track every artifact in your model pipeline

Reproducibility at scale

Know where a model comes from end to end, and what is running where.

Dataset versioning

Reliably capture all changes to the data you’re using to train your models. ‍
Save datasets in any system: GCP, AWS, Azure, or even uploaded directly to W&B servers.
Clearly identify what downstream steps in your pipeline were affected by a change in data.

Model management

Capture all of your trained models in one central system.‍
Maintain a clear picture of all the data, preprocessing, and evaluation that was done on each model.
Trace back the lineage of any production model to the exact code and data that it was trained on.

Key insights from artifacts

What data was this production model trained on?
How did relabeling a chunk of data affect the accuracy of this model?
This data was corrupted — what downstream models were affected by this issue?
How did changing this preprocessing step affect model accuracy?

Explore the Weights & Biases platform

W&B for research & education

Pending...

See how W&B integrates with the ML ecosystem

Regardless of your preferred framework, environment, or workflow, our platform integrates with your existing tools and can be flexibly layered on top of any infrastructure.

Scott McClellan, Senior Director of Data Science and MLOps
NVIDIA

“Putting AI into production requires enterprises to manage a broad range of processes, including data governance, experiment tracking, workload management, and compute orchestration. Pairing W&B with Run:ai MLOps software with NVIDIA accelerated systems and NVIDIA AI Enterprise software enables enterprises to effectively deploy intelligent applications that solve real-world business challenges.”

Multi-cloud and flexible deployment

Weights & Biases offers unmatched cloud flexibility, with secure enterprise deployments via a W&B-managed SaaS cloud or dedicated private cloud.

We are available on all 3 major cloud marketplaces, with full integrations with AWS Sagemaker and GCP Vertex. W&B is also framework and library agnostic, working with popular libraries like Azure OpenAI, Optimized Pytorch and HuggingFace on cloud platforms.

Secure enterprise deployments

Your infrastructure of choice

The world’s leading ML teams trust W&B

Loved by the world’s cutting-edge ML teams.

Stay connected with the ML community

Working on machine learning projects? We’re bringing together ML practitioners from across industry and academia.

Community

Join our community of machine learning practitioners.

Podcast

Go behind the scenes with ML industry leaders.

Webinars

YouTube

Watch videos about cool ML projects, interviews, W&B tips, and much more!

Collaborate Centrally

With the whole team on the same page, you can iterate on models faster. The model registry gives you a central source of truth: it’s clear what model is in production and how new candidate models compare.

Governance at Scale

Gain visibility into the model development process with just a few lines of code. Easily view model lineage and answer questions like “What was the exact version of the dataset this model trained on?”

Transparent

We cite all the algorithms that we’re using, and show all the logs of sweep progress. You have complete visibility.

Lightweight

It’s easy to get started. We’ve dealt with edge cases, so you don’t have to worry about concurrent runs and crashing runs.

Powerful

Our sweeps are infinitely customizable. You can pick your own distribution for inputs, specify logic, and use early stopping.

Parameter importance

Visualize which hyperparameters affect the metrics you care about. W&B comes with default visualizations that make it easy to get started without writing custom code to compare machine learning experiments.

Collaborate Centrally

Visualize all your hyperparameter sweeps in one unified place with our dashboard.

Report Gallery

Authors use Weights & Biases to track and visualize their machine learning experiments, and write reports to explain and showcase their work.

OpenAI Jukebox

Exploring generative models that create music based on raw audio

Digital Lighting Effects

Use RGB-space geometry to generate digital painting lighting effects

Political Advertising

Fuzzy string search (or binary matching) on entity names from receipt PDFs

Visualize Model Predictions

Visualize images, videos, audio, tables, HTML, metrics, 3d objects, point clouds

Driver's View

Semantic segmentation for scene parsing on Berkeley Deep Drive 100K

Debug performance in real time

See how your model is performing and identify problem areas during training. We support rich media including images, video, audio, and 3D objects.

COVID-19 main protease in complex N3 (left) and COVID-19 main protease in complex with Z31792168 (right) from “Visualizing Molecular Structure with Weights & Biases” by Nicholas Bardy

Dataset versioning with deduplication 100GB free storage

Automatically version logged datasets, with diffing and deduplication handled by Weights & Biases, behind the scenes.

The Science of Debugging with W&B Reports

By Sarah Jane of Latent Space

Automatically version logged datasets, with diffing and deduplication handled by Weights & Biases, behind the scenes.

Seamlessly share progress across projects

Manage team projects with a lightweight system of record. It’s easy to hand off projects when every experiment is automatically well documented and saved centrally.

Learn more

About us

Our start

Our values

Honesty

Curiosity

Gumption

Grit

Our leadership

Lukas Biewald

Chris Van Pelt

Shawn Lewis

Yan-David Erlich

Cameron Kinloch

Phil Gurbacki

Lavanya Shukla​

Adrian Swanberg

Our team

Our opportunities

Admin

Admin - Other

Customer Success

Success ML Engineer - Enterprise

Marketing

Marketing

R&D

Engineering

Product

Security

Sales Administration and Business Development

Revenue Operations

Our investors

Our offices

Germany

U.S. Headquarters

Japan

What is your current MLOps Maturity?

Try Weights & Biases

Build vs. Buy

In-House Solution

Weights & Biases

ROI of ML projects

ROI of ML projects

Just a few of the teams on W&B Saas Cloud

Explore the Weights & Biases platform

The Weights & Biases end-to-end AI developer platform

The Weights & Biases platform helps you streamline your workflow from end to end

Models

Weave

Core

The best machine learning teams in the world use Weights & Biases​

Trusted by some of the most innovative teams in machine learning

A single shared source of truth

Integrates with the tools and frameworks you’re using already

Production-grade MLOps at scale

Trusted by 400,000+ ML practitioners

Rapid deployment across teams

Easily access the compute you need, with none of the complexity

Tap into a rich partner network

MLOps Whitepaper

Academic research teams

Managed W&B Private Cloud

Why use Weights & Biases?

1

Automatic Tracking

2

Model Debugging

3

Collaboration

4

Reproducibility

Lineage & Tracking

Storage

Artifacts

Custom plans available.

Key Values of Weights & Biases

Reliable, automatic experiment tracking

Persistent knowledge base

Machine learning team management

Production model lineage

Lavanya Shukla

The best machine learning teams in the world use Weights & Biases

Key Values of
Weights & Biases