Tide / Signal development framework
Your own foundational research lab for financial time-series data. Tide turns your data and objective into a live trading signal, trained in a sandbox only you can see.
Isolated per client
Auditable + versioned
You own the weights
The problem
Foundation models don’t learn from markets.
General-purpose models
Building it in-house means
Data
Fragmented feeds across venues and vendors, aligned from nanoseconds to quarters.
Talent
High competition for ML talent, plus a scientist to lead the lab.
Time
Research is slow and error-prone. Every firm rebuilds the same stack.
Tide gives you the stack, the model and the workflow out-of-the-box, so your team can focus on the data and hypothesis.
Research / Our model vs. foundation models
Built for market data.
Access research
39M-parameter architecture trained from scratch on market data and benchmarked out-of-sample against foundation models up to 8x its size.
GOOG TAQ
Forecast error by horizon
Foundation models lose to a naive “repeat the last minute” forecast. Our model beats both, and is the only model to beat HAR at every volatility horizon.
Intraday volume · rolling
Rolling RMSE vs. the best foundation model (our model vs. Chronos-2).
Intraday volume · at the open
One-shot RMSE for the full day vs. the best foundation model (our model vs. TimesFM 3.0).
Realized volatility · 60 min
Higher out-of-sample R² than the best foundation model (TimesFM 3.0), with no historical-volatility inputs.
TIDE / experiment to deployment pipeline
Data goes in, and a model you own comes out.
Built on our proprietary model architecture
Multi-frequency, retrained from scratch on client input data and modeling objective.
Research engagements
Start with a question.
End with a live signal.
1
Define problem statement
Scope objective, data, and success criteria
2
Develop adaptations
Configure and iterate on the framework
3
Present results
Report, scoring script, and out-of-sample evaluation
4
Go live
Deploy on a live data feed for trading
CASE STUDY
Intraday forecasting
Improving execution.
DATASET
Mixed basket of equities, conditioned on broader market data
ENGAGEMENT
Fixed-fee PoC, optional extension
OUTCOME
Beats every baseline on every equity tested.
DELIVERY
Accuracy benchmarking, execution impact, and path to deployment
DATA ACCESS Residency
Third-party licensed data via Fintech Sandbox, no client data transfer necessary
Scope a PoC
Philosophy
Modeling markets at the level of individual orders.
We founded Lake Shore Dynamics on our conviction that it is necessary to model the market from the level of individual orders. The behavior of market participants is the ultimate filter for the information involved in price and market formation. If we can explain markets at the resolution of order flow, we can realistically assess strategies and model the impact of macroeconomic events and company-specific news on our ability to trade successfully.
We are researchers and builders with backgrounds in mathematics, statistics and computer science. We believe large-scale market data and architectural innovation will capture the idiosyncratic behavior of market participants without imposing biasing assumptions.
Philosophy / In practice
Compressing the order book
Access research
A static projection of the limit order book, compressed to a third of its size, used to forecast one-minute cumulative volume out of sample.
Book depth · NASDAQ ITCH
Price levels per side at 10ms, reconstructed from raw order flow.
Compression · static projection
Smaller book representation, so the model can view ten minutes of rich context.
Evaluation · held-out data
Out-of-sample, forward-looking results, with a standalone scoring script.
Static Limit Order Book representation
More context across space and time with the same richness of information.
The team
Building the future of signal research.
Daniel Gold
Co-Founder & CEO
MSc Computational & Applied Mathematics · University of Chicago · Constellation Software
Christopher Ley
Founding ML Engineer
PhD Electrical Engineering · Continental · Center for Mathematical Modelling
Andrew Dennehy
Co-founder & Scientific Advisor
PhD Candidate, Computational & Applied Mathematics · University of Chicago · NSF Fellow · Leidos
Aleksandr Kovalev
Lead Engineer
MSc Mathematics & Computer Science · Booking.com
Mani Salahmand
Software Engineer
BSc Applied Computer Science · Embedded systems
Get started
It starts with a research question.
Book a demo