Building a research watch list
A finance student uses the daily reports as a reference set for a dissertation on market volatility. The archived logs give a dated, consistent record of model reasoning that can be cited and reviewed over a full term.
Coral Rendova applies risk-filtered AI models to crypto markets and compiles the results into one daily report. No forums, no group chats, no guesswork.
Built for UK students. Designed around risk reduction, not speculation.
Student investors are rarely short of opinions. Price predictions circulate in group chats within minutes of any market move, and most of them carry no analysis behind them.
Coral Rendova removes the noise by processing market data computationally rather than emotionally. The model does not react to sentiment. It reacts to volume, volatility, and historical pattern deviation, then scores each asset against a fixed risk threshold before anything reaches a report.
The model ingests pricing, volume, and order-book data continuously. It identifies pattern shifts before they become visible on a standard chart, giving a measurable head start rather than a prediction of certainty.
Every asset is scored against a volatility threshold before it can appear in a recommendation. Assets that exceed the threshold are excluded automatically, regardless of short-term momentum.
Each report is dated, archived, and written in plain language. You see what the model flagged, why it was flagged, and how the previous day's selections performed.
Coral Rendova was built around a simple constraint: students should be able to understand why a recommendation exists, not just trust that it does. Every output traces back to a documented step in the risk-filtering process, which is described in full on the methodology below.
The platform does not promise returns. It reduces the time required to read a volatile market and makes the reasoning behind each decision visible.
Each report follows the same sequence. No step is skipped, and no manual override changes the order.
Pricing, volume, and liquidity data are pulled from multiple exchanges at regular intervals throughout the day, giving the model a current view rather than a delayed snapshot.
A machine learning model scores each asset against a fixed volatility threshold. Assets above the threshold are removed from consideration before any recommendation is drafted.
Remaining assets are compiled into a single, dated report with plain-language reasoning, published at the same time each morning.
A finance student uses the daily reports as a reference set for a dissertation on market volatility. The archived logs give a dated, consistent record of model reasoning that can be cited and reviewed over a full term.
An engineering student with a modest monthly allowance uses the volatility filter to avoid assets prone to sudden swings. The report replaces hourly price-checking with a single five-minute read each morning.
Below is a sample of the report format. Figures illustrate layout only and are not a record of live performance.
| Date | Assets reviewed | Flagged as high-risk | Report status |
|---|---|---|---|
| Mon | 128 | 41 | Published 07:00 |
| Tue | 131 | 37 | Published 07:00 |
| Wed | 126 | 44 | Published 07:00 |
Sample report format for illustration. Live dashboards are available to registered users.
Every published report is time-stamped and stored in a permanent archive, so past reasoning can be checked against later outcomes. The log is not edited retroactively.
Create an account to receive the next daily report and see how the risk-filtering process applies to current market conditions.
See why students choose Coral Rendova