Private, non-commercial project — for personal use only, not a product or service offered to the public.
Observe · Validate · Execute · Review

The trading engine that
shows its work.

Falconn trades NIFTY systematically and records every decision it makes — each signal stamped with the rule that produced it, every session replayable tick by tick, every loss explained.

One traceable loop — the bench stays separate from the money.
desk.falconn.app — Performance SYSTEM · TRADING
Monitor
Ops
Positions
Analytics
Review
Performance
Journal
Charts
Replay
Research
Research
Strategies
Workbench
Net P&L (validation)
₹5.61L
Trades
106
Win rate
48%
Max drawdown
₹15.3k
₹0₹2L₹4LJul 25Jan 26Jul 26
Cumulative P&L per trade — three forward-validation runs · 106 trades · Jan 2025 → Aug 2026 · net of costs · not live performance
Latest decisions
NIFTY 24500 CE +38.2
STRAT_003 · rule R3.2 · 6 lots
Signal — STRAT_006
vetoed: low-volatility regime
Stop trailed 24 481 → 24 512
100%
of decisions recorded & traceable
8
strategies forward-validated before going live
Tick
level session replay, print by print
1-click
halt — hard risk limits enforced in code
The platform

Six rooms, one process

Everything the engine does lives on one desk — from the live session to the research bench.

📓

Automated Journal

Every signal, order, fill and exit is journaled as it happens — nothing typed in after the fact, nothing forgotten.

Market Replay

Play any archived session back print by print and watch exactly what the engine saw when it pulled the trigger.

🧪

Research bench

Backtests, walk-forwards and paper runs on unseen sessions — strategies earn live orders, they're never granted them.

🔍

Failure-mode analysis

Losses are classified, not mourned: false breakout, weak entry, low-vol regime — with the why on every trade.

📊

Performance

Session-over-session equity, R-multiples, MAE/MFE excursions and exit-reason attribution across every arm.

🛡️

Risk guardrails

Hard order caps, daily loss budgets, per-class position limits and a two-step kill switch — discipline in code, not willpower.

Journal

Every decision, stamped and searchable

The journal writes itself while the engine trades. Each entry carries the exact rule and structure that produced it — so weeks later, one click answers "why did we take this?"

  • Signal → decision → execution → outcome, on one card
  • Strategy version recorded on every trade
  • R-multiple, MAE/MFE and slippage computed for you
  • Notes, tags and grades for the human layer
Trade detail — NIFTY 24500 CE12 Aug · 09:47
StructureHH → HL → breakout
StrategySTRAT_003 · v17
Entry / Stop ruleR3.2 / S2.1
Fill vs signal+0.8 pt
Outcome+1.21R · MFE +1.63R
REPLAY DECISION →
Market Replay — 14 Aug sessionprior day in frame · 1× · 09:40
13 Aug — prior session (static)14 Aug
15:35
Replay

Rewind the market. Watch the decision.

Any archived session plays back print by print, on its own clock. Jump to any trade from the journal and land at the exact moment the signal armed.

  • Tick-level tape, not just candles
  • Pan and zoom mid-playback
  • Deep-linked from every journal trade
  • The same data the engine actually saw
Research

Losses become lessons, automatically

The bench runs backtests and forward validation on sessions the strategy has never seen — and when live trades lose, a deterministic classifier explains how: the taxonomy your next improvement comes from.

  • Forward-validated on unseen sessions before any live order
  • Every losing trade tagged with a failure mode
  • Strategy versions frozen and linked to their validation runs
  • The bench stays separate from the money
Failure modes — last 90 days21 losses
False breakout 38%
Weak / late entry 24%
Low-volatility regime 19%
Time-exit loss 14%
Execution / slippage 5%
One decision, end to end

Every trade travels the same traceable path

Recorded at every step. Reviewable forever. Replayable from signal to fill.

Market
live ticks
Structure
highs · lows · breaks
Strategy
validated rule fires
Risk
limits & sizing
Order
placed
Fill
executed
Outcome
R · MAE · MFE · logged

Get in touch

Falconn is a private project — but questions and conversations are welcome. Leave a note and we'll get back to you.

hello@falconn.app
Built on structure.
We'll only use your details to reply. No spam, no lists.