Beta — under test. AI Alpha Engine is in beta and is being actively tested. Features, data and pages may change or be unavailable, and paid plans are not on sale yet. Nothing here is investment advice.
How AI Alpha Engine Works
AI Alpha Engine is an automated investment research platform with one objective: find the stocks most likely to beat the market over the next one to six months. The order matters, the market is analysed before your portfolio is, and this page follows it.
What is AI Alpha Engine?
The platform automatically collects comprehensive market data: prices, fundamentals, news, catalysts, analyst actions and other investment evidence. The foundation is deterministic. Code collects the data, computes momentum and relative strength against a global benchmark, attributes each piece of evidence to the company it genuinely concerns, and collapses repeats of the same story. Those steps give the same answers from the same inputs every time. That is measurement, not stock selection.
The investment analysis itself is done by AI: models reason twice, at two different scopes. First across the whole covered universe, with no portfolio in the picture. AI analyses the companies, the market, its major drivers and the opportunities across the universe, and decides which stocks are most likely to outperform. That global research is published every market day in the model portfolio, so it is visible without an account. If you provide a portfolio, AI then evaluates your holdings against those opportunities and determines where your capital has the strongest potential to outperform. Both stages cite the evidence they used, which is why the platform can tell you what supports a conclusion.
Deterministic systems provide the data and the measurements. AI performs the analysis and makes the investment recommendations.
How it works
Three steps, and the first one is already running. It does not wait for you to sign up.
Step 11
upload_fileThe engine researches the market
Every market night, across the whole covered universe: market data in, relative strength computed, evidence attributed to the right companies, and then reasoned about, universe-wide, by AI that decides what is leading and which stocks are most likely to outperform. All of it happens without you, and it is published every market day in the model portfolio.
→
Step 22
analyticsYou add your holdings
Import a holdings export from your broker. Your positions and weights become the lens the existing research is read through. They do not start the research off.
→
Step 33
summarizeYou get research about your portfolio
A cockpit summary and a full written report: where your capital has the strongest potential to outperform, what is still supported, and what is weakening. Refreshed every market night, with the evidence cited.
“Every market night” means the night after a trading session. The research is built while the market is closed and is waiting before it opens again. Markets do not trade at weekends, so neither does the engine.
What you receive
dashboardCockpit
A concise summary of the most important signals, risks and review priorities, at a glance.
descriptionReport
Detailed portfolio analysis covering themes, economic drivers, catalysts, and highlighted signals and risk flags.
account_treeDrivers
Portfolio positioning by economic driver: which drivers you hold, how each ranks on relative strength versus the benchmark, and how much of the portfolio sits in the current market leader.
fact_checkThesis
Validation of whether each holding still supports the investment case in light of recent news, leadership shifts, and analyst actions.
newspaperNews
Catalysts, analyst upgrades and downgrades, and market events affecting your specific holdings and watched themes.
workHoldings
Current holdings and economic-driver exposure for each position.
historyPerformance
Compare each recorded portfolio snapshot with today’s prices to see how the original holdings would have performed if left unchanged. This is a hypothetical comparison, not actual account performance.
The analysis in detail
Optional reading. Expand a section to see how the nightly analysis is produced.
The methodology
Five analytical principles underpin every portfolio review, from where capital is flowing to where the evidence is strongest.
publicEconomic Drivers
Markets are moved by a small number of macro forces at any given time: AI infrastructure, energy transition, defence spending, housing cycles. We identify which drivers are gaining dominance and find the leading positions within each.
hubThemes
Every driver breaks down into the themes that carry it: data-centre power and cooling, grid build-out, defence electronics. Themes are where capital concentrates inside a driver, and we track which of them are leading right now.
leaderboardLeadership Detection
Relative strength and momentum, updated each market night, measure where institutional capital is going. The measurement is an input to the analysis, not a verdict. Past winners are never assumed to be future winners, and leadership is re-evaluated from scratch each session.
trending_upLeadership Shifts
The engine tracks where momentum is building and where it is fading, surfacing tickers gaining strength and those where thesis support may be weakening, so changes show up early rather than late.
insightsConviction Signals
Conviction is a judgement, not a formula. AI weighs signal strength, leadership consistency and the quality of the evidence to say where a case is strongest for further investigation and where risk flags warrant closer review.
Example: how an economic driver breaks down
Economic Driver
Electrification Capex
→
Theme
Grid & Power Equipment
→
Holdings
Vertiv, SPX Technologies
From data to decision support
Deterministic measurement, then AI analysis across the whole market, then that same analysis applied to the holdings you own.
Stage 11
cloud_downloadIngest
Comprehensive market data, prices, news, analyst actions, fundamentals and earnings, pulled from multiple sources each market night and normalised into one consistent dataset.
Momentum, relative strength vs the global benchmark, theme alignment and catalyst signals are computed across the full universe. These are measurements, not a selection.
AI reads each company’s measurements, fundamentals, news and catalysts against the market backdrop and judges its expected alpha. The universe is ordered on those judgements, on where the evidence is strongest, not on sector balance.
Stage 44
manage_searchReview
AI reads your holdings against that research: what still earns its place, what is weakening, and where your capital could do better elsewhere.
The analysis and its recommendations, written up with the evidence behind them, so you can check the case yourself.
From signals to review priorities
Every market night the engine surfaces a clear set of signals for each position. The signals are evidence, not conclusions. AI weighs them against the market backdrop to decide which positions still have the strongest case, so you can make your own informed decisions.
Join the community
Discuss portfolio strategy, market leadership, and how to get the most out of AI Alpha Engine with other members.
Join Discord ↗Choose the access level that fits you
It is not for you if you want trade signals to follow, automated execution, or someone else’s conviction. The engine does research and shows its work; the decisions stay yours.
Explore the available plans to see which portfolio analysis, reports and features are included.
View plansAll analysis is generated by AI based on market data. Nothing on this platform constitutes personalised financial advice.
← ← Back to Home