AION Signal Infrastructure
The signal backend AION runs on Indian derivatives, leased to institutions and platforms as an API. Your algorithm on our stack, or the signal families that currently measure best, each with a full signals report. Pricing on request. Not offered to individuals.
The problem
A desk or a platform that wants its own intraday signals on NSE, BSE and MCX derivatives has to build and keep running a stack that has little to do with its edge: a tick feed that does not drop, instrument and lot-size registries that stay current, detectors that run on every tick, a lifecycle for each signal from entry to exit, and an audit store that records what every signal did so the strategy can be judged on evidence rather than memory.
Most of that work is repeated at every firm, and most of it is undocumented once the person who built it leaves. The signals themselves are the small part.
What is leased
The stack AION operates for its own research, as it exists today, delivered to you over an API under a lease agreement. You do not receive the code and you do not host it. We run it, and your systems read from it.
- Tick ingestion. Live market data for NSE, BSE and MCX derivatives through a broker feed, normalised and enriched before any detector sees it.
- Pattern detection. A set of detectors that run on every tick across the covered instruments and raise a signal with an entry reference, a stop, a target and an expected move.
- Signal lifecycle. Every raised signal is tracked to one of three ends: the target reached, the stop reached, or a time exit. Nothing is left open and nothing is dropped from the record.
- Outcome audit store. Every completed signal is written with its result, exit reason, holding time and adverse excursion, and kept for ninety days. The signals report is produced from this store.
- Reporting. Per family and per exchange: signal count, trading days, the share that ended above entry, the tracked result per lot, the exit mix, the average holding time and the adverse excursion.
The detectors, the thresholds and the way the lifecycle is scored are not described on this page and are not shared outside an engagement. The report describes what the signals did, not how they are found.
Two ways to engage
The families that currently measure best
AION tracks every signal family it runs against the audit store. Families that finish the measurement window with a positive tracked result, on a sample large enough to mean something, are the ones offered. The ones that do not are named in the report as well, so you can see the whole picture, not the flattering part.
Your algorithm on our stack
You bring the rule set. We implement it as a detector on the same ingestion, lifecycle and audit store, and it is reported to you with the same metrics as our own families, so your rule set is judged on the same evidence base from its first day.
The signals report
Each engagement starts with the report for the families on offer, and each lease carries it on as a standing deliverable. It is produced from the audit store, not from a backtest, and it includes the misses.
| What it states | How it is measured |
|---|---|
| Signal count and trading days | Completed signals in the window, by family and exchange, and the number of sessions they came from. |
| Share ending above entry | Signals whose tracked exit was better than their entry reference, as a share of all completed signals. |
| Tracked result per lot | Exit reference less entry reference, times one exchange lot, summed and averaged. Tracked, not executed: no order was placed, and no cost or slippage is deducted. |
| Exit mix | The share of signals that ended at target, at stop, or at the time exit. |
| Holding time and adverse excursion | Average minutes from entry to exit, and the average worst move against the signal before it closed. |
| Day consistency | How many sessions in the window the family finished positive, so one large day cannot carry a weak family. |
Measured state on 15 October 2026. Between 16 September and 14 October 2026, twenty trading sessions, the audit store completed 63,300 tracked signals on 1,232 instruments across NSE, BSE and MCX derivatives. Three signal families finished that window with a positive tracked result on a sample above two thousand signals each. The other families did not, and the report says so. The figures themselves are shared with prospective lessees under confidentiality, not published here.
How to use it
- Status
- Available on request to institutions, brokers, fintech platforms and proprietary desks through direct engagement rather than self-service signup. There is no public launch date and no self-service signup.
- Delivery
- API only, under a lease agreement. No code transfer, no on-premise copy, no dashboard for end users.
- Pricing
- On request. The lease is priced per engagement, by the families or the custom work involved and the volume read.
- Not offered to
- Individual customers. AION does not publish these signals, does not sell them to retail users and does not place orders for anyone.
- Enquiries
- [email protected]
What it is not
- It is not investment advice and not a recommendation to buy or sell anything. The lessee decides what, if anything, to do with a signal.
- It is not an execution service. The stack raises and tracks signals; it does not route orders. The Order Execution Agent is a separate product and records rather than decides.
- It is not a backtest. The report is built from signals tracked live on the day they fired.
- A tracked result is not a realised one. No order was placed, and costs, slippage and fills are not in the figures.
What this is not
Informational and educational only. Not investment advice, not a recommendation to buy or sell, and not a SEBI-registered Research Analyst or Investment Adviser product. AION is an infrastructure provider: we publish information and leave the decision entirely with you.
We publish what a system does and what it produces. We do not publish how it is built. Research partners and institutional collaborators can request a technical briefing under confidentiality: [email protected]