So I built my own — a confluence scanner with a risk engine, a signal bot whose backtester refuses to go live on a negative edge. Now I build them for clients: scanners, alert bots, journal automation, private AI assistants.
$ edge status scanner · running — BTC ETH SOL · 15m/1h/4h risk · armed — breaker dormant (0/20 resolved) journal · 1 open, 0 resolved verdict · edge unmeasured — keep logging $ ▌
Five systems I made for my own trading and workflow. Proof I ship real software — not products for sale. Each one listed with its least flattering number.
Multi-timeframe confluence scanner with a Layer-2 risk engine, circuit breaker, Telegram alerts, and a live mission-control dashboard.
Crypto signals on 1-hour bars: EMA200 trend gate, ATR trailing exits, and a built-in backtester that reports R-expectancy.
Polymarket scanner. Alerts only when its model and the market price disagree by 12+ points at real liquidity.
Long-term ETF advisor: VTI/VXUS/BND core, 200-day trend overlay, DCA and threshold rebalancing.
Local-first AI assistant: four specialized agents, vector memory, RAG knowledge base, voice. Runs 100% offline.
* Unflattering numbers stay up. A backtester's job isn't to confirm your idea — it's to kill it before the market does.
Everything I build is yours — code handed over in full, no black-box SaaS you depend on forever.
Market scanners, signal bots with Telegram alerts, trade-journal automation, backtesters that refuse a negative edge. Built around how you actually trade.
The copy-paste around your judgment calls, automated: research, reporting, data extraction — running on a schedule, delivered to Telegram, Slack, or email.
An assistant that knows your business, your docs, your workflows — running locally or on your server. Your data never leaves.
You walk me through how you do it today. I tell you what's buildable, what it costs, and what I'd skip. If it's not worth building, I'll say that on the call.
Scope and price on paper within 24 hours. The price holds for 30 days. Nothing starts until you've approved it and the deposit is in.
Weekly updates while I build. You see it running before the final 50% is due. Then full code handoff — it's yours forever, no subscription to me.
No — and walk away from anyone who says yes. What I build is your rules, executed without you staring at charts, plus a backtest that tells you honestly whether those rules have an edge. If the numbers say no, you find out for the cost of a build, not the cost of an account.
Not yet, and I say that out loud — my journal holds one open trade and zero resolved, so the edge isn't even measured. That's the point of the tooling: find out cheaply, on paper, instead of finding out with size. I sell the measurement, not the outcome. If someone tells you their bot prints money, ask for the journal.
Fair. Everything I've shipped is on this page — that's the track record I have. The structure covers the rest: scope in writing before code, half up front, and you see it working before the second half is due.
Describe the workflow you want automated. I'll reply within 24 hours with a fixed quote.
I'll read it properly and reply within 24 hours — with a number, not a sales pitch.