This Entrepreneur Raised $50 Million To Build Data Tools And Infrastructure For Artificial Intelligence Development

In the world of AI, few founders have navigated the intersection of coding, corporate experience, and entrepreneurial vision as effectively as Ulrik Stig Hansen.

As the co-founder of Encord, Ulrik’s path—from a small town outside of Copenhagen to the bustling tech scene of London and, ultimately, Y Combinator—offers a fascinating roadmap for founders looking to build and scale in the rapidly evolving world of artificial intelligence.

In this interview, Ulrik talks in detail about finding the product-market fit early on, planning how to scale the company, and the motions he went through to make it work.

Listen to the full podcast episode and review the transcript here.

Coding Beginnings in a Small Danish Town

Growing up in a small town outside of Copenhagen, Ulrik had the kind of quiet environment that fostered a deep focus on problem-solving and building. He considers the lack of distractions a conducive environment for aspiring AI founders.

Unlike many of today’s founders surrounded by a buzzing startup culture from day one, Ulrik spent most of five years immersed in coding from his bedroom. He spent time working on different projects and perfecting his skill sets.

Ulrik’s early start in building products and talking to users laid the foundation for what would eventually become Encord. He loved combining problem-solving and building aspects, spending days pondering problems and building things people could implement quickly.

“Coding is unique because you can build things from your bedroom,” Ulrik says. “That ability to turn problems into products that people can use, and to do it quickly, hooked me from the start.”

This passion for building led Ulrik down an atypical path for an aspiring tech founder. Before diving into AI, he sought experience in business and finance since he wanted to start a company someday. He moved to London to join JP Morgan in his early 20s and began his career.

Ulrik recalls how he enjoyed his stint at JP Morgan and learned a lot. However, the corporate world couldn’t quell his desire to build. After a few years, Ulrik left the structured environment of JP Morgan to return to his true passion: building products, this time in the field of AI.

The AI Revolution and Meeting His Co-Founder

The next significant chapter in Ulrik’s journey took place at Imperial College London, where he pursued his master’s in computer science. This was during a transformative period for AI—the debut of transformer models, the rise of GPT-2, and YOLO models.

As Ulrik puts it, AI was “starting to deliver value in the real world for the first time.” Seeing the tremendous potential of AI technologies, Ulrik felt he couldn’t miss this opportunity, which he describes as standing on the precipice of a significant technological platform shift.

Around this time, Ulrik met his co-founder, Eric, in an entrepreneur network in London. Eric brought in a strong background in high-frequency trading and big data systems. He had been putting thousands of models into production to do different things like trade stock futures and ETFs.

Ulrik and Eric quickly realized they had a shared vision. While models and computing were rapidly advancing, the data layer of AI development stacks—particularly in terms of annotation and quality—lagged. This recognition of a gap in the market set the stage for Encord.

Encord: Solving Data Problems in AI

Encord’s original focus was on data annotation, a critical component for AI teams to train their models effectively. Through conversations with over 1,000 AI engineers, researchers, and data scientists, Ulrik and Eric discovered a common pain point.

They discovered the challenge of getting high-quality data annotated for AI models. Over time, the problem shifted from data quantity to data quality. Instead of getting high amounts of data, people were more focused on getting the right data into their AI production systems.

Thus, Ulrik and Erik built the Encord platform that manages the full lifecycle of data management, annotation, and model evaluation–from solving the data quantity to the data quality problem.

Encord effectively breaks down the problem with three distinct pieces of data management creation–getting the right data annotation, having humans provide feedback to the model, and then model evaluation.

Encord’s platform allows AI teams to solve the dual challenges of managing large quantities of data while ensuring that the right data is fed into AI systems. This is crucial because bad data can “poison” models, degrading performance.

Y Combinator: Learning to Scale

In the winter of 2021, Ulrik and his team joined Y Combinator, the famed startup accelerator. This was during the height of the COVID-19 pandemic, so their YC experience was fully remote.

Despite the lack of in-person interactions, Ulrik found immense value in the YC network, particularly as a first-time founder navigating the complex world of fundraising, product-market fit, and scaling.

“For a first-time founder, you probably can’t find a better accelerator than YC,” Ulrik says. The program helped him and his team avoid common mistakes, particularly in areas like finding product-market fit, acquiring initial customers, and scaling out the production engineering teams.

YC also provided a robust network for fundraising, which was particularly valuable as Encord raised its pre-seed and Series A rounds in 2021 during one of the hottest funding environments in recent memory.

Even so, Ulrik maintains that founders must work hard to get the investors and early capital they want. It’s a common misconception that fundraising is easy for an AI company, but this is not the case.

The Fundraising Journey: $50M Raised

Encord’s fundraising journey has been nothing short of remarkable. Between 2021 and 2023, the company raised $50M across several rounds, including a recent $30M Series B led by N47, with participation from Y Combinator, CRV, and other notable investors.

Storytelling is everything that Ulrik Stig Hansen was able to master. The key is capturing the essence of what you are doing in 15 to 20 slides. For a winning deck, take a look at the pitch deck template created by Silicon Valley legend Peter Thiel (see it here), where the most critical slides are highlighted.

Remember to unlock the pitch deck template that founders worldwide are using to raise millions below.

However, Ulrik emphasizes that raising capital is far from guaranteed, especially in today’s challenging environment. “The Series B market is still reeling from the boom years, and it was definitely the most difficult round to raise,” Ulrik admits.

The funds raised have allowed Encord to build out its product engineering teams, build products that it could sell repeatedly to its customers, scale its go-to-market operations, and make some early-stage mistakes that come with finding product-market fit.

But as Ulrik notes, the journey from early-stage to growth-stage brings its own challenges, particularly when transitioning from a vision-driven early-stage company to a metrics-driven growth-stage company.

As Ulrik explains, back in 2021, during the pre-ChatGPT days, people were only peripherally interested in AI. The main focus was on crypto and blockchain, which were hot among investors. However, Ulrik and Eric had a few positives going for them.

Incredible Fundraising Success

For starters, they had an insanely compelling pitch at a time when the AI boom was ready to take off. Encord had already acquired its initial set of customers and had worked with some of the world’s leading AI teams, both at big enterprise models like AI Scaleups.

Since Ulrik and Erich had customer focus and had built the product to support use cases, they could find a set of investors who believed in that vision and decided to back them early on when the team was still small.

Encord raised its pre-seed round in February–the Y Combinator check. It closed the seed round in May and the Series A in early October. Although the typical interval between funding cycles is 18 to 24 months, Ulrik and Erik went ahead with their fundraising.

They had some amazing customers and logos on board early on and knew exactly what they wanted to build. Fundraising was about accelerating the product roadmap to create the pieces of the product that they wanted. Cash gave them the flexibility to hire teams and move faster.

Ulrik explains that they had to consider different stats and metrics like CAC, payback periods, account executive ramp times, efficiency metrics, gross margins, and more when they reached the series B growth stage.

These numbers became more crucial for the later-stage funding rounds than in the early rounds, where 90% of the pitch was about vision and only 10% was about metrics. In the later stages, like series C and beyond, metrics and vision factors are at 50%:50% and up to 30%:70%.

Scaling: The Next Frontier

One of the most significant lessons Ulrik has learned is that scaling a company can be even more difficult than finding a product-market fit. “People don’t talk enough about scaling,” he says. “The air gets thinner the further you go.” Founders don’t have many peers from whom to ask questions.

Encord’s challenge is finding qualified go-to-market talent to sell a technical product in a relatively new market. In other words, it must find high-quality salespeople who can help run the sales functions and deals autonomously.

This has been one of their most significant hurdles as the company has scaled by over 100% in the past two years in terms of employee headcount. To manage this growth, Ulrik has adopted a lean approach to hiring, focusing on bringing in top-tier talent rather than simply expanding headcount.

“More people mean more operational overhang,” he notes, highlighting the importance of maintaining agility in a fast-growing company.

Encord offers higher salaries and equity to bring in a highly concentrated set of highly qualified and talented people who can help accelerate the company’s growth.

Building a Strong Board and Navigating Challenges

As Encord has grown, so too has its board. It had a small board for much of the company’s early history. Still, with Series B, the board expanded to include more perspectives, adding credibility and fresh insights to Encord’s decision-making process.

However, Ulrik cautions that choosing the right board members is critical, likening it to a marriage you can’t easily exit. “You have to make sure you can work with these people for many years.” CRB, Encord’s series A and seed investor, has been on the board and propelled it to where it is today.

Problem-solving and rallying the team around challenges are other areas in which Ulrik remains deeply involved. As a leader, he believes in diving into the details of the most important problems while delegating less critical tasks.

“The hardest part of scaling is maintaining focus and alignment,” Ulrik says.

The Vision for Encord’s Future

When asked what the future holds for Encord, Ulrik’s answer is rooted in his passion for building. “If I woke up tomorrow and Encord’s vision was fully realized, we would be the go-to platform for every AI team in the world, solving the hardest problems in data management, quality, and annotation.”

With a clear vision, a solid team, and a growing market for AI-driven solutions, Encord is well on its way to achieving that goal. Its initial vision was to automate the data annotation process for computer vision, but now it has grown to several new dimensions, ready to solve new problems.

Ulrik’s journey has shown that the path to building and scaling a successful company is filled with challenges, lessons, and opportunities—and he wouldn’t have it any other way. He advises aspiring entrepreneurs not to wait for the right opportunity but to dive right into it.

Ulrik recommends that founders scale and grow with the company and upgrade their skills by leveraging resources and advisors and surrounding themselves with intelligent people.

Meeting the right people and being in social circles where they can openly discuss the various problems and challenges in the business is crucial. These strategies and not being afraid to ask for help can also promote personal growth.

Listen to the full podcast episode to know more, including:

• Early coding experience: Ulrik’s early passion for coding set the foundation for his entrepreneurial journey.
• AI’s potential: The rise of AI models, such as GPT-2, signaled a technological shift Ulrik didn’t want to miss.
• YC impact: Y Combinator helped avoid early mistakes and accelerated product-market fit.
• Fundraising lessons: Raising $50M across multiple rounds required vision, metrics, and perseverance.
• Product-market fit: True product-market fit became evident when the sales team closed deals autonomously.
• Scaling challenges: Scaling Encord involved hiring top-tier talent and navigating a lean, efficient growth strategy.
• Focus on problem-solving: Deep involvement in solving key problems while delegating less important tasks was key to Encord’s success.

 

SUBSCRIBE ON:

For a winning deck, see the commentary on a pitch deck from an Uber competitor that has raised over $400M (see it here). 

Remember to unlock for free the pitch deck template that is being used by founders around the world to raise millions below.