trading systems

0DTE options · gamma advisory · real-time data warehouse

Three coordinated projects: a data warehouse that ingests 429 API calls every 30 seconds, a signal evaluator that scores 62 features per ticker per cycle, and an execution engine with 20+ safety gates. Built in three weeks end-to-end. 2,015 tests across the three repos.

Scope: paper trading on IBKR · personal, not a fund · infrastructure validation over strategy backtest.

── pipeline ─ signal to execution

                                 ingest                              signal                                   execution
                          ┌──────────────────┐              ┌──────────────────┐                    ┌──────────────────┐
GEXBot REST (11 ep/tick) ─│                  │              │                  │                    │                  │
                          │ market-data-     │─── ATTACH ──▶│  gex-advisor     │─── alert_log.db ──▶│  paisamaker      │──▶ IBKR
GEXBot WS (5 hubs)       ─│ fetcher          │   DATABASE   │                  │                    │                  │   paper
                          │                  │   read-only  │                  │                    │                  │   DUP975149
                          │ ● 429 req/30s   │              │ ● 30s poll loop │                    │ ● 24/7 systemd  │
                          │   3-ts provenance │              │   62 feat/tick    │                    │   6s reconnect    │
                          │   22 GB SQLite    │              │   5-force align   │                    │   tolerant        │
                          └──────────────────┘              └──────────────────┘                    └──────────────────┘
                                    │                                  │                                        │
                                    ▼                                  ▼                                        ▼
                           scalar_readings                CASCADE_WATCH   (71% wr)                     discord_signals.db
                           chain_snapshots                CHARM_SQUEEZE   (75% wr)                     (16 tables · WAL)
                           instrument_registry            STRUCTURE_BREAK (filtered)                    EOD flatten 15:45 ET

       ────────────────────────────── shared VM: Hetzner CPX31 · 4 vCPU · 8 GB · Ubuntu 24.04 · $18/mo ──────────────────────────────

       safety plane:   kill switch (L0/L1/L2)   spread gate >30%   bid validation   contract validation   fresh quote re-fetch
                        paper-port fail-closed   exit escalation    EOD flatten       max hold time         position dedup (UNIQUE)

● operational ● degraded — cross-database ATTACH on SQLite WAL, freshness filter data_age_s < 120

── subsystems ─ in execution order

svc-paisamaker[TRADING]uptime 100%

paisamaker · automated 0DTE execution

Python 3.12 · ib_insync · Hetzner VM · 18,387 LOC

Two independent signal sources (a custom GEX evaluator and an external advisor service) feed a unified execution dispatcher. The dual-lane design came from one specific observation: IBKR gateway reconnects take about six seconds, and a single-loop design would miss the next 30-second signal cycle. Separate lanes, independent cooldowns, per-ticker gates. The mode state machine (DATA_ONLY → DEGRADED → WARMUP → FULLY_LIVE) blocks trades during recovery instead of retrying synchronously.

signals / day~100

tests1,327

tickers8

best trade+$410

Python 3.12ib_insyncasyncioSQLite WALsystemdDocker (IB Gateway)

design note: paisamaker-dual-lane →

svc-gex-advisor[POLLING]uptime 99.8%

gex-advisor · SPX gamma exposure signal

Python 3.11 · Hetzner VM · ~11,100 LOC

Three alert types built on a five-force alignment model and a conviction ladder. CASCADE (negative gamma acceleration) hits 71% on SPX 0DTE paper trades across 14 days. CHARM SQUEEZE (afternoon theta decay) hits 75%. STRUCTURE BREAK (level break) is filtered down to noise. Every alert is logged with a full feature snapshot plus outcome fields (MFE, MAE, target_hit, stop_hit) — so new signals ship as shadows first, validate against data, then promote.

tests463

alerts (14d)46

features / cycle62

CASCADE win rate71%

Python 3.11SQLite WALpandasDiscord webhookssystemdStreamlit

svc-market-data-fetcher[INGESTING]uptime 99.95%

market-data-fetcher · real-time data warehouse

Python 3.11 · stdlib + asyncio · 8,387 LOC

429 concurrent API calls every 30 seconds against 39 tickers. A three-timestamp provenance model (event_at, fetched_at, stored_at) lets downstream systems filter on freshness with confidence. REST and WebSocket paths run in parallel with failover; exponential backoff into a circuit-breaker into self-kill + systemd restart. No frameworks, no ORM — stdlib sqlite3 and aiohttp only. Framework-free isn't a badge, it's a choice so every millisecond in the fetch path is auditable.

rps (steady)429

tests225

test runtime2.7s

db size22 GB

Python 3.11asyncioaiohttpSQLite WALprotobufzstdsystemd

── safety gates ─ 20 gates, every trade path

/opt/paisamaker-app/safety/20 gates · all tested

gate	category	fail behavior	tested by
kill-switch L0	kill-switch	system running, entries allowed	test_gex_watch
kill-switch L1	kill-switch	halt entries, exits still run	test_gex_watch
kill-switch L2	kill-switch	flatten all + no new entries	test_gex_watch
max positions global	pre-trade	reject entry, log reason=max_positions	test_signal_dispatcher
max positions per-ticker	pre-trade	reject entry, log reason=ticker_limit	test_signal_dispatcher
max daily loss	pre-trade	halt entries for the day	test_gex_watch
paper-port fail-closed	pre-trade	reject if IBKR port ≠ 4002	test_gex_ibkr_mode
contract validation	pre-trade	reject if 0DTE expiry or class mismatch	test_contract_map
spread gate	pre-trade	reject if bid-ask spread > 30%	test_gex_strike_selector
bid > 0 gate	pre-trade	reject into empty markets	test_gex_strike_selector
fresh quote re-fetch	execution	re-validate drift + spread at submit	test_executor
IBKR mode state machine	state	DATA_ONLY → DEGRADED → WARMUP → FULLY_LIVE	test_gex_ibkr_mode
post-resume warmup	state	60s stability before trading resumes	test_gex_ibkr_mode
position dedup (UNIQUE)	execution	DB rejects duplicate contract_id rows	test_gex_db
idempotent advisor alerts	data	UNIQUE on advisor_alert_id	test_advisor_poller
EOD flatten	exit	force close 15:45 ET	test_exit_lifecycle
EOD panic window	exit	MKT escalation from 15:35 ET	test_exit_lifecycle
max hold time	exit	90m SPX / 60m QQQ / 30m singles	test_exit_lifecycle
exit escalation (3-tier)	exit	LMT@mid → LMT@bid → MKT	test_exit_lifecycle
per-source cooldown	pre-trade	gexwatch doesn't block advisor	test_signal_dispatcher

Every gate returns a specific reject reason that's logged to discord_signals.db.gex_execution_log. No silent failures. Audit by reading the log.

── planned ─ requires read-only VM API

A live alert feed and live trade feed are planned next. Both require a read-only HTTP API on the Hetzner VM that exposes alert_log.db and gex_execution_log to the public internet. About four hours of FastAPI + systemd config, pending a safer deployment posture (rate limits, read-only contracts enforced at the server layer, not trusted from the client).

Live alert feed — tail of the last N alerts with outcome fields
Live trade feed (market hours) — current positions + today's fills
Week-by-week P&L + fill-rate charts
Shadow signal log — signals being validated before promotion

── eof ─