Systematic trading technology company, Kraków, Poland, founded 2024
We build and operate our own systematic trading platform.
Curson Capital is a systematic trading technology company in Kraków, Poland. We believe a disciplined process, applied consistently across a broad universe of equities, is a sounder basis for investment decisions than judgement exercised under time pressure — so we built one and we run it ourselves.
Our platform, Curson, forms return forecasts across several hundred US large-cap names before the market opens, constructs a risk-controlled long and short portfolio from them, and implements it in the market. Every trading day, the same way, unattended. Every decision it makes comes from a rule we can point to and replay.
Execution runs in simulation while we build the track record that precedes launch.
One trading day
times in US EasternFive stages around the New York session, in the same order every trading day since July 2026.
- after the closeUniverse and data
The day's prices, fundamentals and macro context are collected across the universe.
- overnightSignals
Dozens of families of signals are recomputed for every name.
- before the openReturn forecasts
A cross-sectional forecast for every name, with candidate models measured alongside.
- before the openPortfolio construction
Forecasts become the day's target portfolio, once the risk gates pass.
- after the openImplementation
The target is compared with holdings as the broker reports them, and only the difference is traded.
Our approach
One process, five stages, once a day. Breadth rather than conviction, and the same decision every time the inputs look the same.
Universe and data. The investable universe is defined from liquidity and listing criteria rather than from a view. Prices, corporate actions, dividends, fundamentals and macro context are collected daily and kept so that any past day can be reconstructed exactly as it was known at the time.
Signals and return forecasts. Dozens of families of signals are computed across the universe every night, spanning price behaviour, valuation and quality, market sensitivity and macro conditions. They are combined into a cross-sectional forecast for every name before the open, and a model in production keeps its place only until a candidate outperforms it on data neither has seen.
Portfolio construction and implementation. Forecasts become a portfolio rather than a list of ideas: positions sized within limits per name and in aggregate, long and short exposure balanced with a market-neutrality objective. We then trade the difference, not the portfolio, behind a chain of checks that has to pass in full before an order is sent.
Risk control and oversight. Diversification is the first control and the limits are inputs to construction, so a portfolio that would breach them is never built. If the data behind a day is incomplete the process holds cash rather than trading a partial view, every stage records what it decided, and a failure raises an alert within minutes.
- Universe
- US large caps
- Style
- Long / short
- Rebalance
- Daily
- Discretion
- None
The team
Curson Capital was founded in July 2024 by two engineers who had spent years building data and machine-learning systems inside banks, insurers and manufacturers. Research and portfolio construction on one side, infrastructure and implementation on the other. The firm is bootstrapped and has no outside investors.
How the company makes money
Two revenue lines from software we own, in the order we intend to open them.
First, proprietary trading. We trade the platform with the founders' own capital, and the returns are the revenue. It needs no customers, no sales cycle and no permission — only a track record good enough to stake our own money on, which is exactly what the platform is accumulating now.
Then, subscription access. Self-directed investors get the platform's research output as a service: the daily scores, the target book and the evaluation reports that show how the model is holding up. Recurring digital revenue from software we already operate, with a marginal cost close to zero. It opens when the track record earns it, and prices are published here before anyone is charged.
Funded by us. Bootstrapped and founder-funded. No outside investment. Our cost base is cloud infrastructure, market data and the brokerage connection — no offices, no inventory, no billable hours.
Shipped recently
Flow failures now reach Slack within minutesInfrastructure
Any orchestrated flow that fails, crashes or times out posts to the monitoring channel automatically. Until now a failed morning run could go unnoticed until someone looked at the dashboard; now the alert arrives before the market opens.
Feature build moved to the previous eveningPlatform
For two weeks in August the morning feature build occasionally overlapped with inference, inference saw a partial cross-section, and the book correctly fell back to cash. The technical features are now computed the evening before, and every downstream step is chained on the completion of the one before it rather than on a clock.
Balanced long and short book goes livePlatform
The balanced long and short book became the only portfolio the platform trades. Shorting sits behind an explicit switch, and a book containing a side the platform is not cleared to trade stops the whole run rather than trading half of it.
Get in touch
Questions about the platform, the company or the way we work all reach the two founders directly. We reply to every email, usually within two business days.
Or write to either founder:
- hieronim.kubica@cursoncapital.com — Hieronim Kubica, PhD, co-founder and ceo
- michal.szarek@cursoncapital.com — Michał Szarek, co-founder and coo
Curson Capital is based in Kraków, Poland. There is no contact form on this site and no tracking: an email is the whole channel.
Nothing on this site is an offer to sell or a solicitation to buy any security, a recommendation, or personalised investment advice. No performance figures, live or simulated, are published here.