Product / Docs
Product docs
Start with the Python library: choose a universe, use a signal, build a portfolio, run a local draft, then send a verified backtest. The docs explain each step.
Python library
from alphora import Alphora
client = Alphora(api_key=os.environ["ALPHORA_API_KEY"])
universe = client.universes.crypto_perps_top_20()
premium = client.features.premium_reversion(universe)
portfolio = client.construct.market_neutral(premium)
draft = client.backtest_local(portfolio, start="2022-01-01")
run = client.backtest(portfolio, start="2022-01-01")
run.paper_trade()
Getting started
Install the Python package, authenticate once, pick a universe, compose a strategy, test it locally, then send a verified backtest.
Library reference
Functions for universes, signals, portfolio construction, risk controls, and both local and verified backtests.
Composition guide
How you or your AI combine reusable signals into supported strategies with known structure and reproducible results.
Backtest semantics
What `backtest_local()` is for, what `backtest()` guarantees, and why the verified run is the one that matters.
API preview
A lower-level API for teams that want the same signals and run model without the Python wrapper. It is not the main path at launch.
AI agent guide
Rules for Cursor, Claude, or other agents so they stay inside supported Alphora workflows.
Answer-first guides
These pages are designed for the questions people ask before they are ready to read a full API or library reference.
Research process
Understand the practical difference between backtesting and paper trading, what each one is good for, and why serious strategy work needs both.
Tool intuition
Learn how to interpret Monte Carlo equity paths, which metrics matter, and how Alphora's GBM simulator is useful without pretending to be a full backtest.
Tool intuition
Learn why cross-strategy correlation matters, how it affects portfolio behavior, and how Alphora's multi-strategy simulator helps make the tradeoffs visible.
Research process
Learn why out-of-sample performance matters, what it reveals that in-sample cannot, and how Alphora's product framing keeps that distinction visible.
Research process
Learn a practical research workflow for Hyperliquid HIP-4 markets, from data modeling to forward validation.
Research process
Learn how do you evaluate a carry trade before calling it market neutral, why it matters in carry and basis trading, and what to validate before trusting the workflow in live research.
Research process
Learn how do you turn a volatility forecast into a trade filter, why it matters in volatility forecasting and dispersion, and what to validate before trusting the workflow in live research.
Research process
Learn how do you size uncorrelated sleeves, why it matters in portfolio construction and risk budgets, and what to validate before trusting the workflow in live research.
Research process
Learn how do you spot backtest leakage, why it matters in backtest design and evaluation, and what to validate before trusting the workflow in live research.
Research process
Learn what walk-forward testing is, why traders use it instead of one clean split, and how it fits into an honest validation workflow.
Research process
Learn how do you audit a dataset for strategy research, why it matters in data quality and microstructure, and what to validate before trusting the workflow in live research.
Research process
Learn how to backtest a binary signal strategy honestly, including trigger rules, holding windows, rebalance behavior, and the trading costs that often dominate threshold systems.
Research process
Learn how do you monitor a paper strategy day to day, why it matters in paper trading and live monitoring, and what to validate before trusting the workflow in live research.
Research process
Learn how to think about news and catalysts in Hyperliquid HIP-4 markets without reducing every event to a generic sentiment trade.
Research process
Learn how to paper trade a Hyperliquid HIP-4 idea so you can test the contract logic, timing, and execution assumptions before going live.
Research process
Learn how rebalancing interacts differently with binary and continuous signals, and why cadence, thresholding, and turnover policy belong to the same design decision.
Research process
Learn how should you read cross-strategy correlation output, why it matters in quant tool intuition, and what to validate before trusting the workflow in live research.
Research process
Learn how should you interpret time to expiry in HIP-4 markets, why it matters in hyperliquid HIP-4 markets, and what to validate before trusting the workflow in live research.
Research process
Learn how should you think about volatility forecasting for HIP-4 markets, why it matters in hyperliquid HIP-4 markets, and what to validate before trusting the workflow in live research.
Research process
Learn how to compare scheduled and event-driven rebalancing, when each approach fits better, and what to validate before choosing one.
Research process
Learn how do you turn a raw score into a position size, why it matters in signal and portfolio design, and what to validate before trusting the workflow in live research.
Research process
Learn how threshold bands reduce overtrading, when they help, and what to validate before trusting them in a live strategy.
Research process
Learn how do you compare a backtest against a real baseline, why it matters in research workflow, and what to validate before trusting the workflow in live research.
Research process
Learn how execution assumptions distort backtests, which shortcuts do the most damage, and how to validate a strategy more honestly.
Research process
Learn how partial and full rebalances differ, why traders choose one over the other, and what the trade-off really is.
Research process
Learn how do you combine multiple signals in one portfolio, why it matters in signal and portfolio design, and what to validate before trusting the workflow in live research.
Research process
Learn how do you use a simulator before building a strategy, why it matters in quant tool intuition, and what to validate before trusting the workflow in live research.
Research process
Learn how turnover caps affect portfolio construction and why cost-aware allocations often look different from frictionless ones.
Research process
Learn how do you size a HIP-4 trade, why it matters in hyperliquid HIP-4 markets, and what to validate before trusting the workflow in live research.
Research process
Learn how trading probability calibration works, what it should be measured against, and why a sharp-looking score is not enough on its own.
Research process
Learn why a strong-looking signal can still fail the tradeability test and what to check before mistaking prediction quality for a live strategy.
Research process
Learn how hysteresis bands differ from simple thresholds, why they reduce chatter, and what to test before trusting them in live trading.
Research process
Learn how to set a trading threshold when costs are nonlinear and why one clean global cutoff often fails in live execution.
Research process
Learn why an apparently calibrated trading probability can drift live and what to monitor before the failure gets expensive.
Research process
Learn how do you get an AI agent to compose a trading strategy without inventing unsupported logic, why it matters in aI agent strategy composition, and what to validate before trusting the workflow in live research.
Research process
Learn how to walk-forward test calibration and threshold rules so they do not borrow credibility from one full-sample fit.
Research process
Learn how do you turn a market idea into a machine-readable strategy spec, why it matters in aI trading prompts and strategy specs, and what to validate before trusting the workflow in live research.
Research process
Learn how do you set up an automated trading strategy from scratch, why it matters in automated strategy setup and first rollout, and what to validate before trusting the workflow in live research.
Research process
Learn how do you keep an automated strategy simple enough to debug, why it matters in automated strategy setup and first rollout, and what to validate before trusting the workflow in live research.
Research process
Learn how do you automate backtests without losing reproducibility, why it matters in backtest automation and experiment pipelines, and what to validate before trusting the workflow in live research.
Research process
Learn what makes a regime filter robust, which failure modes usually expose overfitting, and how to pressure-test context rules before trusting them.
Research process
Learn how to test whether a regime or context overlay adds real value, and what evidence should exist before it earns a place in the live stack.
Research process
Learn how do you automate paper trading before you promote a strategy further, why it matters in paper trading automation and rollout readiness, and what to validate before trusting the workflow in live research.
Research process
Learn how to evaluate whether a regime filter generalizes across assets and cycles, and what kinds of degradation are acceptable versus disqualifying.
Research process
Learn how do you design alerts that catch real strategy problems instead of creating noise, why it matters in strategy monitoring, alerts, and kill switches, and what to validate before trusting the workflow in live research.
Research process
Learn how to tell whether a context overlay adds independent information or is just a more complicated version of the main signal.
Research process
Learn when a regime filter has become maintenance theater, and how to decide whether to retire, simplify, or replace it.
Research process
Learn how do you turn a score into a ranked book, why it matters in cross-sectional ranking and selection, and what to validate before trusting the workflow in live research.
Research process
Learn how do you turn a trading signal into an execution workflow, why it matters in execution workflows and order automation, and what to validate before trusting the workflow in live research.
Research process
Learn how do you keep automated execution from breaking a valid strategy, why it matters in execution workflows and order automation, and what to validate before trusting the workflow in live research.
Research process
Learn how do you decide what should trigger a trading workflow, why it matters in event-driven triggers and scheduled jobs, and what to validate before trusting the workflow in live research.
Research process
Learn how should you use AI to generate trading ideas without trusting it blindly, why it matters in aI research copilots and signal discovery, and what to validate before trusting the workflow in live research.
Research process
Learn how do you turn an AI-generated trading idea into a testable hypothesis, why it matters in aI research copilots and signal discovery, and what to validate before trusting the workflow in live research.
Research process
Learn when should you roll back an automated strategy rollout, why it matters in rollout environments, staging, and rollbacks, and what to validate before trusting the workflow in live research.
Research process
Learn how do you stage a trading workflow without duplicating the whole system, why it matters in rollout environments, staging, and rollbacks, and what to validate before trusting the workflow in live research.
Research process
Learn how do you rebalance a ranked strategy without turning it into churn, why it matters in cross-sectional ranking and selection, and what to validate before trusting the workflow in live research.
Research process
Learn how do you automate a portfolio made of multiple strategy sleeves, why it matters in portfolio automation and sleeve orchestration, and what to validate before trusting the workflow in live research.
Research process
Learn how do you keep portfolio automation from hiding which sleeve is actually failing, why it matters in portfolio automation and sleeve orchestration, and what to validate before trusting the workflow in live research.
Research process
Learn how do you audit why an automated strategy made a trade, why it matters in strategy state, memory, and decision audits, and what to validate before trusting the workflow in live research.
Research process
Learn how do you design manual overrides for an automated strategy, why it matters in human review, approvals, and override flows, and what to validate before trusting the workflow in live research.
Research process
Learn how do you plug a custom AI signal into an Alphora strategy, why it matters in custom signals and AI tooling integration, and what to validate before trusting the workflow in live research.
Research process
Learn how do you keep an AI trading agent from using future data by accident, why it matters in point-in-time data and agent safety, and what to validate before trusting the workflow in live research.
Python
from alphora import Alphora
client = Alphora(api_key=os.environ["ALPHORA_API_KEY"])
universe = client.universes.crypto_perps_top_20()
premium = client.features.premium_reversion(universe)
portfolio = client.construct.market_neutral(premium)
draft = client.backtest_local(portfolio, start="2022-01-01")
run = client.backtest(portfolio, start="2022-01-01")
run.paper_trade()
Planned API shape
const res = await fetch(
"https://api.alphora.io/v1/features/crypto/premium-reversion?as_of=2026-04-30&universe=crypto-perps-top20",
{ headers: { Authorization: `Bearer ${process.env.ALPHORA_API_KEY}` } }
);
const panel = await res.json();Launch preview
People should be able to see exactly how a strategy moves from code to verified backtest to paper tracking, and where the planned lower-level API fits if they need it.
One launch email. No funnel.