3 Things we learned building Tribe and why project-based work will change AI
- How we’re building an alternate path for top machine learning engineers and data scientists at Tribe
- What we’ve learned about attracting and retaining top tech talent and running successful AI projects at scale
- Why the future of ML will be project based
In 2019, we started Tribe AI, a community of top engineers and data leaders that partner with companies to solve real-world problems with AI. In just over two years, we’ve grown to 150 machine learning engineers, we’re running dozens of projects, and we’ve paid out millions of dollars to our community.
But first, let’s back up and explain how we got here.
The power of an alternate path
Tribe started as an experiment. We met at South Park Commons (SPC), a technical community in San Francisco, where we were surrounded by world class machine learning engineers who had stepped away from big tech. Working at Google, CapitalG and Gigster, we had seen companies of all sizes struggle to recruit ML talent. So we wanted to know: what did this group care about so much they’d be willing to forego the high-paying jobs that most engineers only dreamed about?
The answer was simple: freedom.
Freedom to stop climbing the corporate ladder or to start their own companies. Freedom to do research or see a variety of problems in applied settings across industries while developing innovative solutions. Taking a full time job at any one company was at odds with these goals and not everyone had the financial runway or desire to be a founder. And, while consulting provided freedom, it came with a host of drawbacks – no community, too much busywork, and a lack of interesting ML problems.
For us, the opportunity became clear. We needed to find a way to give them freedom, a community of other engineers, and the infrastructure to lean into their engineering superpowers. We knew that if we could build this alternate path, the talent would come and the customers would follow.
Along the way, we’ve learned a lot about what attracts and retains top talent and what makes a successful AI project. We’ve honed in on three key pillars that have informed everything we do: freedom, community, and specialization. Here’s why:
1. Freedom drives success for everyone – if you have the right infrastructure.
There’s nothing new about the idea of freelancing or the gig economy. But all too often there’s an inverse relationship between freedom and quality. And a perception that (gasp) outsourcing leads to inferior results. To make Tribe successful, we knew we’d have to turn that belief on its head.
So we built Tribe to be a more efficient and higher performing AI lab than any in-house team. Through trial and error, we learned the best way to prioritize and scope projects to quickly deliver value to clients. We learned the nuances of ML product management and how vital this is to producing results. We handled all the sourcing, communications, and project management, so engineers didn’t have to.
2. Community unleashes ML superpowers.
At Tribe, community has been a vital part of how we define freedom work from day one. Most people (even brilliant ones) don’t do the best work of their lives alone. They grow by doing deep work and connecting with others based on mutual interests and passion. The word community gets thrown around a lot, but the reality is there aren’t a lot of places to forge these kinds of connections outside of the classroom or office.
At every step of the way – from building out teams for projects to vetting every single applicant – building a strong, tight community has been our goal. We wanted to give talented, entrepreneurially-minded, machine learning experts a place to find each other, share ideas, and connect. To take on the projects that would unleash their talents and say no to the other ones.
If magic comes from putting brilliant minds in the same room, we’re building Tribe to be that room.
3. Specialization is the future of (better) work.
At Tribe, we believe in the power of specialists. And that means freeing our talent to do the work they’re best at – building models and solving tough problems with data – and none of the other stuff. We’ve invested in the infrastructure to make it possible for all our technical talent to work this way and the results have been impressive.
Over the last two years at Tribe, we’ve built dozens of teams to work across companies of all sizes, industries, and stages. We’ve done projects with a fractional team of exactly the right 3-4 specialists and, in six weeks, achieved more than a full-time team was able to accomplish in six months.
Every project we’ve done since founding Tribe has reinforced our belief that bringing together the right combination of skill sets and deploying them on the right problem is the key to unlocking outsized value.
The future of ML is project based
At Tribe, we set out to build a company that combines the compensation and technical camaraderie of an organization like Google with the autonomy of being a founder. Today we are 150 AI engineers, data architects, and product leaders across practically every dimension of technical expertise. And we get dozens of new applicants every week.
Our work has not only generated outsized results for our customers, but also created economic freedom for our community. This has allowed them to quit jobs, travel the world, volunteer, start companies, and spend more time with family.
To us, it’s clear that project-based work will change the way companies build AI into their businesses. For starters, it’s the way top talent wants to work. And it’s better for companies as well. Until now, top ML talent has only been accessible to a few tech giants. And, even for them, hiring is slow, difficult, and expensive. Under this new model, we’ve seen companies – from legacy enterprises to startups – accelerate ML adoption and see results faster in a way that would have been impossible with traditional hiring.
When we look at the future of Tribe what excites us most is this potential for real world impact across companies of all sizes and industries. This is the future ML – and the future is bright.
Written by By Jaclyn Rice Nelson