🧠 A brain for your database - How is MindsDB democratizing Machine Learning

MindsDB wants to make ML and AI reasonably accessible to companies who can’t otherwise think of doing so. It was a brilliant idea but Rome wasn’t built in a day.
No items found.
Priyanshu Anand
May 17, 2022

Intelligent solutions for an intelligent world

What steam engines were to the first industrial revolution, Machine Learning and AI are to our generation of scientific advancements. Did you know that at the beginning of 2020, the number of bytes in the digital universe was 40 times larger than the number of stars in the observable universe? Google, Facebook, Microsoft, and Amazon alone store at least 1,200 petabytes of data on their global databases.

The amount of data being produced even just by organizations globally is staggering and even incomprehensible. Hence, the explosive innovation of Machine Learning (ML) models and AI has finally allowed enterprises to leverage predictive algorithms for precise decision-making. The global ML market size is expected to grow to $21.17 billion in 2022 as compared to $15.44 billion in 2021. According to a report by Fortune Business Insights, the market is expected to grow to $209.91 billion by 2029 at a CAGR of 38.8%.

Data-driven solutions and technology are powering competitive decision-making across industries too. Predictive analytics is being used extensively by tech giants to drive personalized customer experience. Retail conglomerates are using it to study customer purchase patterns to target more personalized advertising. BFSI sector has leveled up with ML, offering better onboarding, credit scoring, anomaly & fraud detection, and wealth handling. Healthcare has been one of the most impacted industries by using AI in detection and diagnosis through algorithms, some of which have reportedly overperformed humans too.

However, adopting specialized machine learning continues to be expensive for enterprises to overlook the overhead expenses, and somehow justify implementation by bleeding resources. The thing is, even though there’s plenty of data available to most companies, it’s not ‘clean’ enough to be put against the likes of predictive modeling. You need proper expertise which costs a significant budget, patience, and iterations to truly reap any formidable benefits.

Common Machine Learning Challenges - source

So, what can we do about it? Apparently, there’s a way to leverage existing databases and manipulate them to build efficient working models. It’s called AutoML and the entire idea is to be able to surpass operational roadblocks to utilize the miracle of ML and AI for solving real-world problems i.e. find a way to bring these models closer to the data. Treading along the same line of thought - since data is fundamental to machine learning, and SQL is fundamental to databases, what if there was a way to bring the two together?

That’s exactly what MindsDB does. Let’s take a closer look at how they do it, but first, some backstory.

The minds behind MindsDB

Adam Carrigan and Jorge Torres go way back. They studied together at The Australian National University where Adam was doing his Master in Finance and Jorge in Information Systems focusing on Machine Learning. They went on to become really good friends, so much so that they even lived together, worked on various projects together, and took part in university activities together. Parted ways after college to pursue other things for a while. Notably, Adam completed his MPhil from the University of Cambridge and worked as a Management Consultant at Deloitte. Jorge went on to work as a Software Engineer across various companies including Skillshare.

Until finally they were reunited in London, started working on a project, and Co-founded what was called Real Life Analytics. It was an AI-based model for targetted advertising across demographics. It didn’t work. But it did lead to a whole lot of lessons and learnings, one of which was understanding how deeply intricate it was to build working ML algorithms and how not having the budget to pay for an engineer made them realize the challenges of implementing it for smaller, more trivial problems.

This is how MindsDB came to be. Its mission is to help enterprises power innovations by easily connecting databases to machine learning models. They want to make ML and AI reasonably accessible to companies who can’t otherwise think of doing so. It was a brilliant idea but Rome wasn’t built in a day.

There’s the saying that when you know, it’s ready, it’s probably too late. In the beginning, we were waiting for when it was ready to kind of show it to people to go about it. But actually, going back to those very early days, the more we waited, the more dangerous it was, and now we understand that the product is never ready. - Jorge Torres (Meeting of the Minds)

Democratizing Machine Learning

MindsDB brings the capabilities of Machine Learning to SQL databases with something known as AI tables. What an AI table does is integrate machine learning models as tables inside the database, create predictions, and can be queried with simple SQL statements. All seemingly difficult things like classification predictions or optimization recommendations can all be handled simply by running an SQL query. Essentially, anyone with the knowledge of SQL can quickly visualize and deploy ML models.

And it’s open-source. In fact, keeping MindsDB open-source is fundamental to the founders’ vision of democratizing Machine learning for every database user. The thing about ML and AI is that it’s still nascent, which means there’s still a lot of room for experimentation. And with capabilities that make data visualization possible simply through database integrations, training and testing scenarios further extend what the product can achieve.

MindsDB has partnered with the world’s most trusted, flexible, scalable, multi-cloud databases like Confluent, MariaDB, MongoDB, MySQL, PostgreSQL, Snowflake, and more, to have various avenues which MindsDB’s AI-table based predictions can integrate with. Most of these databases are, again, open-source which allows users to make improvements and increase the efficiency of operationalizing machine learning.

Their attempt of giving ‘a brain’ to databases hasn’t gone unnoticed. Adam has voiced his opinion on raising funding and talks about how it’s important to find investors who share the same vision. It’s essential that before they write you a cheque, they understand the product and truly believe in it; there needs to be a mutual fit between you and someone who’s investing not just money but also their trust into your product. Closing their seed round in November 2021, MindsDb announced total funding of $7.6 million from YC, Walden Catalyst Ventures, OpenOcean, and more.

As more open-source database users get hold of machine learning tools, it will be very exciting to see what the community will produce and how we’ll benefit from AI and ML going forward. Now is a good time for every database developer to get their hands dirty and try out with their own data how the MindsDB integration practically works. - Patrik Backman, Founder and General Partner at OpenOcean.

There are a ton of things you can do to be a part of the MindsDB community. As the adoption of ML and AI increases, their implementation becomes more business-friendly. MindsDB's approach to democratizing machine learning is unique, which is what allows them to grow concerning both idea and usability.

Let’s take a look at how users can get involved with their community 👇

Powering Innovation with Community

MindsDB innovation is community-focused. A large reason for their platform being open-source is about encouraging customers to stretch its capabilities. Their community makes up for more than people who use the product, but people who define the future roadmap and the framework that it’s built on.

Contribute to Github

MindsDB’s open source project is on Github with 6.8K stars as of now, and as a user, you can contribute to the Github repository and help steer the framework for building MindsDB. By interacting with their repository, you also get to interact with other contributors directly and maybe even collaborate with them.

Also, take part in discussions - ask questions, post answers, share insights, discuss ideas, best practices, talk about projects, and get feedback. This community is moderated by MindsDB engineers, so being a part gives direct access to talk to the development team.

MindsDB Community on Github

April announcement for the most helpful contributors in the community

Product Updates on the Community

A community on Slack

That’s right, MindsDB also has a Slack community that’s 1000+ members strong as of now. As a part of this community, you can stay up-to-date on announcements, engineering or product feature updates, and discussions on using MindsDB.

Zoran Pandovski making an announcement in the community

The Rewards Program

MindsDB knows how to encourage open source contributors to make more helpful contributions. You can get rewarded for contributions to code, content, and more. The contributions are divided into 3 levels, each rewarding the community for different types and varying levels of contributions -

For community writers or content contributors, to be a part of the Ambassador program, you can get started by publishing tutorials. As a starting point, there’s a project board that details a list of ideas, work under process, and published tutotials. Take a look at the published work get an understanding of what it should look like.

MindsDB encourages community-led contributions and depending on how valuable your contributions have been, you can also get sponsored for you work.

The entire community journey for contributors in the community has a seamless transition from one stage to another. Starting from something as trivial as posting a bug to making content contribution and coming up with great feature ideas - MindsDB has an extremely cohesive strategy for nurturing members. After becoming an ambassador, a community member naturally builds an incentive towards having bigger contributions in the community. MindsDB rewards them by making them community maintainers.

Webinars

MindsDB hosts webinars in collaboration with its partners to highlight some use cases. If you work with data and are somehow connected to delivering actionable insights with ML or AI, these webinars will be of tremendous help. Featuring industry experts, engineers, and data scientists, you can either register for the webinar and also watch it on-demand.

MindsDB webinars

Events

Apart from insightful webinars, they also take part in other events like conferences, summits, or talks. For example - Jorge spoke at the Pecona Live event last year which is the largest global event about open-source databases.

Watch the recording here 👇

Documentation

An inside-out guide to working with MindsDB - from installation and setup to community tutorials, SQL APIs, and creating custom ML models on the database

Being an open-source platform, you can also contribute to the documentation on Github -

How to contribute to MindDB’s documentation

Blog

Detailed articles and papers on machine learning models, databases, and making the best use of intelligent systems to make better, informed decisions.

MindsDB blog

Community is the future of intelligent solutions

For an open-source solution, the possibilities are limitless. And the MindsDB team has understood how empowering a community of product users to make contributions, can exponentially impact the future of the company. And it’s not just about an organization, there’s something much bigger at play here. The idea of making intelligent database solutions accessible simply with a working knowledge of SQL is a giant leap towards making AI and ML capabilities available to everyday enterprise-grade solutions.

MindsDB is on a mission to democratize machine learning and make life easier for developers or those who make data-driven decisions. They realized that a mission that’s larger than life needs more hands-on deck. As the community grows, MindsDB grows with it. They’re very keen and open to any kind of feedback. There’s also an option for developers to become a Beta Tester who -  gets first access to all new features, becomes a vital part of what helps shape the future of the product and can help figure out the problems or recommend improvements to those features.


To summarize

Intelligent solutions for an intelligent world

  • Some stats about data and a scale of how much of it we produce on the regular.
  • The use of that data through ML and AI models.
  • Challenges of adopting Machine learning into everyday tasks or across enterprise-grade solutions.
  • The idea of AutoML

The minds behind MindsDB

  • Adam and Jorge meet in college and quickly become really good friends.
  • Reunite in London and work on their first project - Real Life Analytics.
  • Building MindsDB.

Democratizing Machine Learning

  • What MindsDB really is - a solution that brings ML to your database, making models as easy as querying SQL.
  • Learnings and lessons.
  • Making it open source.
  • Partnering with the world’s most scalable cloud-based databases.
  • Closing Seed round of funding for $7.6 million.

Powering Innovation with Community

  • How can you get involved with MindsDB’s community -
  • Contribute to GitHub - make contributions to the open-source project.
  • A community on Slack - a 1000+ member strong Slack community for MindsDB users.
  • The Rewards Program - 3-level rewards program that gives Swag to helpful contributors in the community.
  • Webinars - events in collaboration with partner integrations to help the community better use MindsDB for their databases.
  • Events - other global events like conferences, summits, and talks.
  • Documentation - complete guide to working with MindsDB and contributing to the open-source documentation.
  • Blog - detailed articles around machine learning and databases.

Community is the future of intelligent solutions

  • Feedback loop and Beta tester program.
  • Roadmap to the future.


About us

Threado is the command centre for your online community. Signup for the waitlist and check out our cool drops - Community OS, which ended up being the #1 Product of the Day on Product Hunt, SamosaBot, which is a fun way of showing gratitude in your slack community, Serendipity Bot, which connects you to new people in the community based on your interests, and Community Jarvis, templates and automation to help navigate community building!

We’ve 4500+ subscribers! Folks from Canva, Webflow, OnDeck, Pitch, Grammarly, Y Combinator, Freshworks, AWS, and more read our newsletters and community-led stories! Don’t forget to subscribe now and join the bandwagon!

Community Hour is back with another episode featuring Sarah Hawk, Chief Operating Officer at Discourse. Join us on the 24th of May 2022 at 10 PM EST.

If you are a community-led organization and would like us to share your story with our subscribers, just holler! 🙂

Explore other apps

No items found.

Explore similar

Case Studies