We help mission-driven organizations harness their data to work smarter and offer more impactful services using data science, machine learning, and AI.
We help organizations around the world tackle challenges in climate change, education, health, society, science, disaster recovery and more.
150+ projects delivered
We organize competitions on our competition platform to crowdsource social impact problems to 100K+ data scientists and engineers.
$4.8M+ prizes awarded
We have popular open source projects, open source the results of our competitions, and can help you open source your own project.
10K+ GitHub stars
DrivenData is an AI firm with over a decade of experience building end-to-end solutions with organizations ranging from tech-enabled nonprofits to Fortune 100 companies. We help from strategy to implementation and R&D to production, often working as a thought partner to define what's worth doing and how to get it done. To learn more, book a free discovery call.
We develop machine learning models, drawing on leading tools and practices in the field. We deliver well-designed, accessible solutions that have real-world impact.
We worked with the World Bank to develop machine learning models using satellite imagery to identify crops and agricultural climate risks in Yemen, informing development programs in the country after years of civil war.
We build solutions that take advantage of modern advances in generative AI and large language models like ChatGPT. We help teams explore what is possible with the latest technology applied to their needs.
We partnered with the National Center for State Courts to develop a private sandbox environment for testing large language models (LLMs) like ChatGPT, supporting secure experimentation with new capabilities in generative AI.
We use best practices in data science to pull together data sources for analysis, drive inquiry based on organizational needs, and synthesize results. We bring the expertise and tools to illuminate the most useful information from data.
We worked with Science for America to analyze data from the Department of Education and develop an interactive data visualization platform and open source code library, supporting greater understanding of trends, patterns, and disparities in STEM education.
Our competition platform channels the skills of 100K+ data scientists and developers around the world to build solutions for social good. Competitions engage a large expert community with real-world problems and elevate the best solutions.
We partnered with Meta AI to run the Hateful Memes Challenge, where thousands of participants around the world competed to build the best machine learning models for identifying hate speech in multimodal memes containing images and text.
We build modern web applications built on best practices in software engineering, data science, and human-centered design.
We worked with EverFree, a leader in supporting survivors of human trafficking, to develop their Freedom Lifemap technology. The platform includes fully-featured functionality to check in with survivors on their current needs and connect them with personalized support, while strengthening organizations with data to guide and improve programs, measure outcomes, and stop exploitation.
We help take machine learning applications from development to production, bridging the “missing middle” between static research code and the ability to use regularly generated model outputs for real user needs.
We partnered with the Max Planck Institute for Evolutionary Anthropology and the Wild Chimpanzee Foundation to take competition-winning models for detecting wildlife from trail camera videos and build them into a production system runnable on new data. The solution included an open-source tool called Zamba (“forest” in Lingala) for identifying hundreds of kinds of animals, and a user-friendly web application called Zamba Cloud for training and running models in the cloud.
We help you turn your project into an open source software package, incorporating conventions and best practices from the open source community.
We built and currently maintain Cookiecutter Data Science, a widely-used (>8,500 stars, >2,500 forks) open source project template for organizing data science work so that it’s easy to reproduce and build upon. See our open source page for additional examples.
We design and implement scalable, efficient data pipelines to process, clean, and transform data for analysis and machine learning. Our pipelines ensure reliable, reproducible workflows tailored to organizational goals.
We partnered with Candid (formerly Foundation Center and GuideStar) to configure pipelines for increasing data quality and matching nonprofits and funders with the correct organizations in their central database. Deploying and monitoring these systems at scale helps increase the reliability of the core information on philanthropic activity that Candid provides for the social sector.
We develop data warehousing solutions that consolidate and structure your data for easy retrieval and analysis, especially by data analysts and less technical staff. Our solutions are tailored to meet performance needs while ensuring scalability and security.
We worked with CodePath to plan and implement their core data infrastructure to answer critical questions on organizational effectiveness, operations, and impact. The infrastructure brought together data streams from across the organizations for accurate, secure, self-service reporting.
We help turn raw data into well-structured datasets prepared for development and use. Our approach builds on extensive experience developing, evaluating, and sharing a wide range of datasets through our data science competitions.
We worked with researchers at BetterUp and the University of Pennsylvania to develop the CANDOR corpus, a large, novel multimodal dataset of naturalistic conversations including audio, video, and transcripts as well as extracted vocal, facial, and semantic features and post-conversation reflections.
We help you understand what is possible with advances in data science and AI, define what is actually worth doing, and scope the right approach for your goals.
We worked with Fair Trade USA to assess opportunities using their data and potential improvements to their data model to track and communicate impact. These recommendations along with initial prototypes of use helped demonstrate the untapped value of existing data and led to investments in a new team with supporting infrastructure for leveling up the use of data across the organization.
We inventory how data, AI, and algorithms are used in target domains and synthesize findings for action through summary reports and technical resources.
We partnered with Wellcome to investigate the risks of bias in the fields of computational neuroscience and psychiatry along with approaches to address them. As part of the work, we created a public annotated case study illustrating best practices for measuring, mitigating, and communicating algorithmic bias.
We help hire, advise and train data teams. We help organizations bring on their first data science and engineering hires, and help upskill data leaders and teams looking to elevate their practices.
We partnered with the technical leadership team at the Harvard Business Review to hire their first data scientist, helping to craft the job description, source candidates, and lead the applied technical interview to identify the best fit for the role.
We lead workshops bringing together stakeholders to brainstorm, design, and prioritize new opportunities for deploying modern data and AI technologies at their organizations.
We worked with the Black Wealth Data Center to run a half-day design workshop on how generative AI could be used to support their goals of analyzing drivers of Black wealth, in particular generating ideas for understanding media narratives around Black leadership and what questions, hypotheses, and data sources could provide perspective with modern AI tools.
Our experts will work with you to understand the problems you face and how we can leverage your data to offer more impactful services.
Detected wildlife in images and videos—automatically and at scale—by building the winning algorithm from a DrivenData competition into an open source python package and a web application running models in the cloud.
Built solutions using LLMs for multiple real-world applications, across tasks including semantic search, summarization, named entity recognition, and multimodal analysis. Work has spanned research on state-of-the-art models tuned for specific use cases to production ready retrieval-augmented AI applications.
Used satellite imagery to identify crop extent, crop types and climate risks to agriculture in Yemen, informing World Bank development programs in the country after years of civil war.
Analyzed millions of mobile money records to uncover patterns in behavior, and then combined these insights with human-centered design to shape new approaches to delivering mobile money to low-income populations in Tanzania.
Built a real-time, interactive map to visualize attacks on the Ukrainian health care system since the Russian invasion began in February of 2022. The map will support partner efforts to provide aid, hold aggressors accountable in court, and increase public awareness.
Automated recognition of agricultural entities (such as crops, pests, diseases, and chemicals) in WhatsApp and Telegram messages among plant doctors, enabling new ways to surface emerging trends and improve science-based guidance for smallholder farmers.
Used machine learning to match students with higher education programs where they are more likely to get in and graduate based on their unique profile, with a focus on backgrounds traditionally less likely to attend college or apply to more competitive programs.
Visualized the flow of fair trade coffee products from the farms where they are grown to the stores where they are sold, connecting the nodes in supply chain transactions and increasing transparency for customers and auditors.
Analyzed repayment behaviors across dozens of pay-as-you-go (PAYG) solar energy companies serving off-grid populations throughout Africa, and developed KPIs to facilitate standardized reporting for PAYG portfolios.
Mapped out the probabilistic patient journeys through hospitals based on tens of thousands of patient experiences, giving hospitals a better view into the timing of the activities in their departments and how they relate to operational efficiency.
Flagged public health risks at restaurants by combining Yelp reviews with open city data on past inspections. An algorithmic approach discovers 25% more violations with the same number of inspections.
Built algorithms that put apples-to-apples labels on school budget line items so that districts understand how their spending stacks up and where they can improve, saving months of manual processing each year.
Aided anti-trafficking efforts at border crossings and airports by combining data across locations and surfacing insights that give interviewers greater intelligence about the right questions to ask and how to direct them.
Translated advances in machine learning research to practical software for clinical settings, building an open source application through a new kind of data challenge.
Developed online, white-label data science competitions for students to synthesize their learnings and test their skills on applied challenges. Each capstone features a real-world dataset that focuses on an important issue in the social sector.
Founded in 2014, we're an expert team of data scientists and engineers interested in doing good work for good reason. DrivenData was started by three grad students out of the Harvard Innovation Lab with a mission to bring the power of data science to organizations tackling the world's biggest social challenges.
Over more than a decade, our team has worked with some of the most prominent organizations in the social, public, and private sectors. We have built on our machine learning competitions with a full range of data science and AI services, along with a top-notch team behind them.
We engineer code so that it's clear, correct, and reproducible. We've seen a lot of data science in the wild, and we know the importance of organizing work so that it's easy to understand and build upon. Our technical team takes pride in bringing the best tools and practices to the job at hand.
Part of our job is connecting the technical capabilities of what we do with why they matter for people and organizations. We often draw on the tools of human-centered design to help shape the opportunities worth solving and to refine solutions with end users.
We take our responsibility as stewards of data seriously. Our team adheres to best practices in data security, have experience working with extremely sensitive datasets, and have developed tools for data scientists to tackle critical discussions around issues like privacy and bias.
We bring a highly client-focused and iterative approach to our work. By interspersing collaborative discussion with agile development sprints, we are able to respond to emerging priorities and drive towards high-value work outputs that provide the greatest return for development resources.
We believe in the power of collective progress. We support open source solutions where possible, actively maintain open source projects for the data science community, and run competitions where technical experts around the world advance solutions for tough problems in AI/ML.
Delivering impactful work means meeting organizations where they are. We bring a rich range of experience to serve the end-to-end needs of our partners, from strategy and R&D to prototyping and production.
Learn about how winners detected cognitive decline using speech recordings and social determinants of health survey data
Overview of key open-source packages for extracting features from voice data to support ML applications
An introduction to using large language models via the benchmark to a document summarization challenge.
Learn about the results and winning methods from the early literacy screening challenge.
Finding data jobs for good causes can be difficult. Learn strategies, job lists, and tips to find organizations with open positions working on causes you care about.
Learn about the winning solutions from the Youth Mental Health Challenge Automated Abstraction and Novel Variables Tracks
Learn about the winners and winning solutions from the final stages of the Water Supply Forecast Rodeo.
This year DrivenData celebrates our 10th birthday! We've spent the past decade working to use data science and AI for social good. Here are some lessons we've learned along the way.
In this guest post from the MIT Gabrieli Lab, we'll show you how to get started with the literacy screening challenge!
Learn how to process text narratives using open-source LLMs for the Youth Mental Health: Automated Abstraction challenge
Learn about the top datasets sourced for Phase 1 of the PREPARE Challenge.
Inspiration and resources for teaching students data science, machine learning, and AI skills with DrivenData competitions.
Learn about the top solutions submitted for the Pose Bowl: Spacecraft Detection and Pose Estimation Challenge.
Learn about the winning models for forecasting seasonal water supply from the first stage of the Water Supply Forecast Rodeo.
Announcing the V2 release of Cookiecutter Data Science, the most widely adopted data science project template.
Key practices from the field of open science for making data science work more transparent, inclusive, and equitable.
Dive into the solutions from the super segmenters who best detected kelp in Landsat imagery!
Meet the winners with the best systems for detecting clinical terms in medical notes.
Learn about the top visuals created for the Pale Blue Dot: Visualization Challenge and the solvers behind them.
Join our newsletter or follow us for the latest on our social impact projects, data science competitions and open source work.
Have a question about how we can work together or something else? Feel free to contact us!
DrivenData does not have any restrictions in terms of size, mission, software tools, or database structure. We have worked on applications in fields from education to climate, tech to healthcare. Our goal is to partner with organizations that share our vision of using data to make the world a better place. If you're an organization excited about innovative ways to leverage your data, please get in touch.
DrivenData is a social enterprise that works with mission-driven organizations to drive change through data science and engineering.
Challenges are one way we work with organizations to develop data science and AI solutions, by structuring problems so that a global community of experts can drive progress. We think it's a great option for suitable problems, and we've spent over a decade building out a competition platform and community that we're proud of.
We also know that competitions aren't right for every problem. For more flexible needs or sensitive data sources, we have our own team of experienced data scientists and engineers to take the case. Our team has extensive experience developing machine learning and artificial intelligence solutions, from exploratory research to advanced modeling to production systems, with a range of partners including Microsoft, Meta AI, NASA, Gates Foundation, and the World Bank.
To join a competition, check out our open competitions or learn more about how it works. You can also explore our past competitions that often include open solutions and open data.
To run a competition, read our information for partners or reach out with information on the competition you have in mind and someone on our team will follow up to discuss.
Engagements are typically structured in one of two ways.
Deliverable-based engagements target a specific set of outputs. This tends to be a good fit when those outputs are clear and the path to achieving them is well understood.
Sprint-based engagements allow our team to work more flexibly through a series of development sprints, iterating with clients and responding to emerging priorities and learnings as we go. This tends to be a good fit when there is a wider set of potentially useful directions or the nature of the work is more iterative (R&D tends to fall in this bucket).
In both cases, we work with our partners to intersperse collaborative discussion with agile development. This allows us to stay closely aligned throughout the work and drive towards high-value work outputs that provide the greatest return for development resources.
Check out our sample projects, or take a look through our case studies.
If you have additional questions about specific domains or technologies you're interested in, drop us a line below.
Contact us below. It's likely someone on our team will be a good fit and interested in following up.
Let us know the challenges your organization faces and where you're looking for help. We'll reply within 2 working days, and can arrange a free discovery call.
You'll get to talk directly with our data science and AI experts. We'll advise on the opportunities we can see to help and which of our services would have the most impact.
