Quantitative Researcher — Crypto Perpetual Futures

Overview

We build and run algorithmic strategies on crypto perpetual futures. You'll own strategy research end to end — from hypothesis and data to a validated backtest to a strategy running on real capital. This is not a support role feeding numbers to someone else: your work is the product.

What you'll do

• Own strategies end to end: hypothesis → data → backtest → validation → live deployment → monitoring.
• Research funding and basis dynamics, market microstructure (order flow imbalance, book pressure, absorption), and cross-venue relationships in perpetual futures.
• Build backtests that survive contact with reality: taker/maker fees, funding payments, slippage, queue position, latency. Gross alpha that dies at the fee line is not alpha.
• Validate against a defined internal standard — full-year coverage, out-of-sample testing, stress scenarios, parameter sensitivity — before anything touches capital.
• Monitor live performance and diagnose divergence between backtest and production.
• Work with our engineers to move validated research into production execution.

What we're looking for

• Strong applied statistics and time-series background.
• Python at a working research level: pandas, numpy.
• Ability to build a backtest and to defend it — its assumptions, its biases, its failure modes.
• Solid grasp of futures and perp mechanics: funding, basis, margin, liquidation.
• Genuine skepticism toward your own results. We care more about someone who can kill their own strategy than someone who can produce a Sharpe of 4 on paper.
• English, written and spoken.

Nice to have

• Order book data experience.
• Backtesting frameworks — framework (Freqtrade, Nautilus etc) or your own.
• Live trading experience, including strategies that failed and why.

Education

A degree in a quantitative field is useful, but what you can demonstrate matters more. If you have research you can walk us through, that outweighs credentials.

What we offer

• Real capital and a clear path to deployment — research here is not an academic exercise.
• Full remote, flexible hours, async-first.
• Direct access to decision-makers. Small team, no layers between research and production.
• Performance-linked compensation on top of base.
• Institutional-grade data and infrastructure.

Process

Intro call → technical conversation on your past research → conversation with CTO → take-home task (a realistic research problem, not a LeetCode exercise) → offer.