Quantitative consultancy

Evidence-driven systems for
trading, research and AI.

OEAlgo works with funds, trading firms and technical teams on the hard, unglamorous parts of quantitative work — designing strategies that survive honest testing, building the research infrastructure behind them, and integrating data and AI into processes that already exist.

Services

Four areas, one discipline

Each engagement is scoped to a concrete outcome and a clear definition of what "working" means before the work starts.

01

Quantitative Trading

Systematic strategy design and portfolio construction — signal research, position sizing, risk limits, execution modelling and realistic cost and capacity assumptions. Built to be run, monitored and reasoned about, not just backtested.

02

Research & Development

Hypothesis-driven research with the guardrails that make results trustworthy: leakage-free data alignment, out-of-sample and subsample testing, parameter sensitivity, and a written record of what was tried and rejected as well as what worked.

03

Data Science

Time-series and panel modelling, statistical analysis, and the data plumbing underneath it — ingestion, validation, point-in-time correctness, and reporting that a decision-maker can actually read.

04

AI Integration

Large language models and agentic workflows applied where they hold up: research assistance, document and data extraction, internal tooling and automation. Scoped against a baseline, with the failure modes named up front.

Approach

How the work is done

  • Evidence over narrative

    A result is reported with its assumptions, its costs, and the tests it failed. Negative findings are delivered as clearly as positive ones — they are usually worth more.

  • Reproducible by construction

    Every number traces back to a script, a dataset version and a run you can repeat. No results that exist only in a notebook someone closed.

  • Realistic from the start

    Transaction costs, slippage, capacity, borrow and data availability are part of the first model, not a disappointment added at the end.

  • Built to hand over

    Documented code, stated invariants and a working environment. The goal is that your team owns the result — not that you keep paying to understand it.

Contact

Let's talk about the problem

A short description of what you're trying to find out, and the constraints around it, is enough to start. Enquiries are answered personally.

info@oealgo.com