TL:DR; We figured out how to fit custom equity factor risk models in seconds instead of weeks. And we can do this across the entire financial analytics landscape. Ad hoc thematic factor construction (think COVID), risk decomposition through time, optimized factor selection, all on the fly and in seconds.
Try out our demo here.
We each spent nearly a decade at BlackRock and Bloomberg, building products, leading research teams, and empathizing with frustrated clients. They sit at the cutting edge of investment research, coming up with market-beating strategies. Yet they have to make do with a 20-year-old analytics toolkit that current vendors offer.
We are changing that. If the AI community can fit a 300 billion parameter model, there’s no reason why a factor risk model should take a week to compute. In fact, it doesn’t. We figured out how to do it live and in seconds, and we can do it across the entire analytics landscape.
Hi, we’re Sebastian and Misha.
Sebastian, CFA built his expertise in quant research during his time as a Director in BlackRock’s Financial Modeling Group where he implemented and researched equity risk models that analyze trillions in assets.
Prior to Bayesline, he was at Bloomberg, where he incubated the next generation of customizable and actionable quant products as part of the Quant & AI Research group.
A computer scientist by training with M.Sc. in Finance and a passion for quant research, Sebastian spent the last 10 years leveraging the power of machine learning to challenge, innovate, and reshape how institutions think about financial modeling.
Misha, PhD was among the youngest Managing Directors at BlackRock. He headed the portfolio risk research team that evolved Aladdin’s portfolio risk models across all asset classes.
He also headed the team that developed Aladdin’s economic scenario engine and investment models that manage roughly $400 billion in strategic asset allocations.
Misha has spent the past 10 years coupling his professional quant training with his personal interest in all things AI and hands-on engineering.
We met in 2016 at BlackRock, and lived together during Covid. We both have a very strong mission to revolutionize financial analytics. We thought that the world’s most renowned institutions would be the best place to do that. But it turns out that they move too slowly and are not ready to truly innovate, especially in the age of AI. Leaving our lucrative careers was the only way to pursue our dreams.
All of financial analytics was built in the 90s on fundamentally an outdated tech stack (C++ on CPUs). The AI community has developed a new stack (Python ML packages on GPUs) that is orders of magnitude faster, both in time-to-market and run-time. Our aim is to rebuild all of financial analytics on this stack; from economic scenario simulation engines for banks and insurers to portfolio risk models for hedge funds. In the future, Bayesline will be integrated in all financial institutions, eliminating the days of waiting for calculation that tens of thousands of industry professionals have to deal with.