The RoleYou will build the research platform behind our AI researcher. The AI researcher implements and tests signals, and you build and own everything it relies on: the backtesting engine, the data pipelines that feed it, the tooling that evaluates results, and the path from research to live trading. You'll work directly with researchers and engineers to decide what the platform should do, and you'll ship without waiting for a perfect spec. The AI researcher runs experiments around the clock, so the backtester is the referee, and a referee has to be fast, correct, and very hard to fool.
As a member of the founding team, you'll directly impact every level of the business, from product strategy to team culture.
What you'll do- Work directly with researchers and engineers to build the platform our AI researcher uses to turn ideas into signals: the signal framework, reusable feature and model components, and experiment tracking
- Build and own the backtesting engine and the research APIs that researchers, engineers, and agents build on: walk-forward simulation, realistic fills and transaction costs, portfolio construction, and results that reproduce exactly
- Build and run the pipelines for our large financial datasets. Ingest messy vendor feeds, validate them, store them point-in-time, and get corporate actions, delistings, and security identity right across US equities. Make every dataset easy to find and hard to misuse
- Develop the evaluation tooling that decides whether an idea is real: signal diagnostics, risk attribution, statistical tests, and automated checks that catch look-ahead, leakage, and bugs in agent-written code before anyone trusts a result
- Ship to production and own what you ship: deployment, monitoring, and reconciliation of live trading against the backtest, in systems where mistakes cost real money
- Profile and speed up heavy numerical workloads so the loop from idea to evidence takes minutes, not hours
What we're looking for- At least 2 years building research or trading systems at a quant fund, trading firm, or systematic investment team, or equivalent depth shown through shipped systems
- Expert-level Python and deep fluency in its numeric stack (NumPy, pandas or Polars, SciPy), with strong engineering fundamentals: testing, debugging, profiling, and writing code other people can build on
- Experience building pipelines over large financial datasets, especially US equities: vendor feeds, point-in-time fundamentals, corporate actions, and symbology
- Hands-on experience building or maintaining a backtester, and a clear understanding of the systematic investment process: signals, risk models, portfolio construction, and transaction costs
- A solid grounding in statistics and working knowledge of machine learning: enough to get the platform's math right and to notice when the numbers don't add up
- Experience working directly with researchers and engineers, building the tools they depend on. You can take a half-formed request, ask the questions that pin it down, and turn it into a system they trust
- You default to shipping. Given ambiguity, you produce a working system and iterate, rather than waiting for requirements to firm up
- Low ego, high standards, fast execution
Compensation- Salary of $150,000/year to $250,000/year
- Generous equity in the company
- Company-subsidized health insurance including medical, vision, and dental
- HSA, FSA, dependent care FSA, and 401(k) plans provided
- Regular wellness, commuter and learning subsidies
