Medusa Investment Partners

We're building the AI-native operating system for lean investment teams: boutique hedge funds, new launches, independent asset managers, and family offices priced out of institutional-grade operations tech.

Medusa is a three-layer platform: ingestion (we meet you where your data lives: Excel, Azure, broker files, Bloomberg), reconciled data layer (one version of truth, produced by AI-driven reconciliation), and AI agent layer (conversational risk, optimizer, automated reporting & RFP/IR workflows, custom agents).

We work on top of your existing stack. No rip-and-replace.

Why We Exist

Tens of thousands of lean investment teams are structurally locked out of institutional operations tech. MSCI / Axioma / Clearwater charge $100K–$500K+/year and require migrating your data onto their platforms. That leaves most boutique funds running investment risk, client reporting, and reconciliation on 10–20 interconnected Excel files, held together with VBA, ad-hoc Python, and human memory, with severe key-person risk, missed breaks, and end-of-month surprises.

We built Medusa because AI finally changed the economics. LLMs made it viable to ingest a fund's messy Excel sprawl without writing custom ETL per client, a capability that didn't exist 18 months ago. Combined with 22 years at the intersection of finance, technology, and data science, that unlocks institutional-grade infrastructure at lean-team price points for the first time.

The Thesis Behind the Product

GenAI is collapsing the long-standing divide between fundamental and quantitative investing. The "translation tax", the hidden cost of humans shuttling between PMs, risk teams, and systems, disappears when LLMs handle semantic bridging natively. Edge shifts from model to context. A new role emerges: the "Context Engineer": part librarian, part system designer, part investor. It is also why Medusa is model-agnostic by design: the underlying model is swappable, and you are never locked to one AI lab.

Large firms (Pictet, Millennium, Citadel) have the resources to staff the engineering Quantamental 2.0 requires. Boutique and mid-size firms don't, which is the specific gap Medusa is built to fill.

Quantamental 2.0 is the thesis. Medusa is the product that makes it reachable for lean teams.

Read the full thesis on our blog →

Why we're different

Hedge-fund-insider-turned-builder

11 years inside hedge funds (7 at Pictet). Built the workflows we're now automating.

AI-native architecture

~10× cost collapse in ingestion and implementation. Institutional-grade tech at lean-team prices becomes viable for the first time.

Built on your stack

Provider-agnostic. Works on Bloomberg / MSCI / Axioma / Enfusion / Excel, whatever you already have. No rip-and-replace.

Your book, as your team sees it

Your sector definitions, your model mix, your conventions. Reconciled underneath, so the numbers hold up.

How we engage

Two parallel tracks, often running simultaneously:

Track B: Forward-deployed engagement (today's entry point)

Founder-led. We ingest your investment data across 5+ sources, automate your highest-pain manual workflow, deploy conversational agents on top.

Track A: Platform subscription (scaling path)

Standard modules (Risk, Reporting & Communications, Reconciliation, Optimizer) available on monthly subscription as your custom work standardizes.

Pricing: From CHF 10K pilot + CHF 3.5K/month platform fee. Replaces $100K–$500K/year of enterprise risk analytics. Pilot fee credits toward Q1 platform fees on conversion. Larger firms: scoped on a fit call.

Aligned incentives: As custom work standardizes, it moves into the platform subscription and the custom retainer declines over time. Our incentives sit with delivering value, not with keeping bespoke work perpetual.

Founder-led, always. Petr is in every engagement personally.

Our Team

Petr Merkuryev | Founder

Petr brings 22 years of experience at the intersection of finance, technology, and data science.

4 years as portfolio manager at a quant equity hedge fund. 7 years at Pictet's hedge fund division, one of Switzerland's most prestigious private banks, running quant research and managing risk on a multi-strategy hedge fund platform.

Author of the Quantamental 2.0 thesis. Conference speaker on AI in Investment Management.

Petr's vision for Medusa: make institutional-grade investment operations tech reachable for the lean teams that were priced out of it for decades.

petr.merkuryev@medusaip.ai linkedin.com/in/merkuryev

Advisors

Sven Bouman

Advisor, Hedge Fund Leadership & Strategy

Sven is an advisor to Privium Fund Management Ltd (UK) and Cura & Senectus Investment AG. He is responsible for Prime-Port, which provides solutions to investment managers looking to undertake fund distribution activities in Europe.

Between 2008 and 2022, he was the CEO of Saemor Capital, which he founded with the backing of insurance company Aegon as a cornerstone investor. Saemor managed a European equities market-neutral long-short hedge fund with a systematic/quantitative approach to investing. Prior to that, he was Head of Equities at Aegon Asset Management in The Netherlands, responsible for over 15bn AuM.

Sven started his career in the financial industry as a PM at ING IM in 1995. He managed several equity funds (Global, Europe, Japan, Emerging Markets and ESG) and advised on multi-asset mandates.

He is a certified financial analyst (CEFA) and certified institutional investment advisor (MiFID II Stay Compliant Program of CFA Society VBA Netherlands). He holds a master's degree in Economics from the University of Amsterdam. His research on seasonalities in stock returns has been published in academic journals including The American Economic Review.

Koye Somefun, PhD

Advisor, Quantitative Research & AI

Koye has more than 25 years of experience in the industry working in multiple quantitative roles within the multi-asset investment space and at multiple locations. He also worked for the Dutch National Institute for Mathematics and Computer Science as a researcher, publishing several articles on automated trading strategies powered by AI algorithms.

Koye holds a PhD in computational economics and a master's degree in applied mathematics from the University of Notre Dame in the US. He completed his undergraduate studies at the Erasmus University in Rotterdam in the Netherlands. In 2007, he obtained the Certificate in Quantitative Finance (CQF) in London. Koye is based in Amsterdam.