The 10x VC: Moonfire's tech-first vision

When I first met Mike back in 2020, he told me about his fascination with transformers, a type of AI model. Fast forward three years and they’re now powering potentially the biggest change since the introduction of the iPhone.

When we started Moonfire, we wanted to build the most quantitative venture capital firm in the world. The vision was to establish a new form of venture, where a team armed with data, algorithms, and in-house software enhances its ability to execute with better knowledge and greater velocity than anyone else in the industry: a bionic VC.

We continue to pursue that vision today. We see ourselves as a tech company whose business is venture, rather than a venture firm that dabbles with tech. And with the takeoff of AI in the last six months or so, it makes it an even more exciting – and vital – space to explore, experiment and excel. With AI at its core, there’s no reason we won’t see the 10x VC.

I want to talk about our vision of what tech-first VC means, how we practically implement it at Moonfire, and what it means for our founders and LPs.

A partnership in predictions

Prediction is the essence of both AI and VC. AI makes predictions by learning patterns from large datasets, finding trends, and extrapolating data to make educated guesses. VCs do much the same, while also..."trusting in their gut".

This shared reliance on forecasting the future makes for a formidable partnership. As AI continues to advance, the potential for its use in venture grows stronger, promising more accurate and efficient investment decisions. It will empower VCs to make better-informed decisions at scale and at speed, significantly outperforming traditional VCs in terms of investment outcomes. On the same road, just in a faster vehicle.

It will give rise to a new class of venture capitalists – the 10x VCs.

Think about hedge funds. In the 1940s, when the first hedge funds emerged, the focus was on using long and short equity strategies to generate returns. Then, in the 1990s, as computer technology advanced, quantitative analysis began to gain popularity in the hedge fund industry. Hedge fund managers began to recognise the potential benefits of using quantitative models and algorithms to analyse data and make investment decisions. Now, hedge funds are synonymous with quants.

That’s where VC is headed with AI. By 2025, more than 75% of VCs will be informed using AI and data analytics, according to Gartner.

But the key to success lies in striking the right balance between human expertise and AI capabilities, ensuring that the predictive power of both fields is harnessed effectively to create a new era of outstanding investment performance – much like how RLHF (reinforcement learning with human feedback) has powered the advances of GPT-4 and co.

The bionic VC

This approach has always been the core of Moonfire – merging human and machine to transform the scale and accuracy of venture.

We integrate technology across the full VC value chain, from sourcing to exit. That means that we’re not just creating a single model and integrating it into a traditional pipeline. We’re building our operations from the ground up and approaching the problem like a software development project.

• We use our own large language models to find new companies that align with our investment philosophy, and help our founders hire and partner with the best people.
• Our automated and ML-powered sourcing and evaluation pipeline enables us to see 2 million companies a year, instead of 10,000, and filter them more efficiently.
• And we’ve run nearly a trillion portfolio simulations to help us – and the VC community – inform the optimal portfolio strategy.
• This tech-first approach also makes us a better partner to our founders. We can not only create tools for them, but also better understand, support and build alongside them.

Our tech stack allows us to both go deep and broaden our reach. It solves the mental-manual work, so we can focus on the sort of work that humans are better at – actually investing in and forming relationships with founders. And it enables more comprehensive coverage to help us find the best-fit opportunities, while also making VC more inclusive: finding talent wherever it is, and finding it early.

This is all the more pressing now, given the sheer scale of the European tech industry. While the market value of European tech fell by $400bn in 2022, from $3.1tn to $2.7tn, and investment slowed, the ecosystem remains resilient. The combined value of UK tech companies alone reached over $1tn by the end of 2022 – making the UK only the third country in the world to pass this milestone after the US and China.

AI will supercharge the tech ecosystem and the way we invest in it. Entrepreneurs will be able to spin up companies faster, with much smaller teams. Working in the earliest stages as we do, where the available data is always limited, AI helps us spot them sooner.

Humachine learning

That same fascination that fired up Mike and me three years ago still fuels our team’s discussions today. From developing automated investment theses to programmatically reducing biases in our investment decisions, we’re busy integrating AI throughout the venture process.

And the beauty of building our AI models and tech stack in-house is that it’s ours to tune. Our Investment Committee meetings, where every single member of our team – investors to engineers to ops – has a say, help us make better decisions and collectively refine our AI pipeline.

So here’s to the rise of the 10x VCs and a new era where human and machine work together to forge and find the next generation of groundbreaking companies.