DEV Community: dyl4488

I Built an AI That Trades Crypto and Options Automatically — Here Are the Real P&L Numbers

dyl4488 — Mon, 16 Mar 2026 14:00:52 +0000

Most AI trading content is vaporware. Here's what actually happened when I gave an AI $215 and full autonomy to trade and operate a real system.

I did not give it a sandbox account and fake screenshots. I gave it real keys, real constraints, and told it to optimize for one thing: net profit after fees. This was not a "write a strategy on paper" experiment. It was live execution with money at risk.

The raw numbers:

Deposited: $215.00
Fees paid: -$7.94
Active positions: ETH, SOL, LINK, BTC
Trade count so far: high enough to feel every bad fill

What the agent built in one sprint was more useful than most Discord signal stacks:

ORB (Opening Range Breakout) options strategy with a backtested +59% return profile.
Fear & Greed DCA module that only scales buys when Fear & Greed is below 15 (historical 30-day win rate around 80%).
A multi-factor signal engine that scores setups before capital is deployed.

Here is the core scoring formula it uses to rank each setup:

score = (
    0.30 * trend_strength +
    0.25 * momentum +
    0.20 * volatility_regime +
    0.15 * liquidity_quality +
    0.10 * news_alignment
)
trade_allowed = score >= 0.70

The signal engine is intentionally boring. No magic. Just weighted factors, a threshold, and strict filtering.

For ORB, the highest-conviction setup came from the 5-minute window (9:30-9:35) instead of the classic 15-minute range. Entry logic looked like this:

opening_high, opening_low = get_range("09:30", "09:35")
if now > "09:35" and price > opening_high and day in ("Mon", "Wed", "Fri"):
    buy_atm_option()
    set_take_profit(+100)  # premium doubles
    set_stop_loss(-50)     # cut premium in half

That rule set is simple enough to automate and strict enough to test honestly.

The painful truth: fees are the enemy. On Coinbase, a 0.6% taker fee is 1.2% round trip. You need +1.2% just to break even. If your edge is tiny, fees erase it. This is exactly why small-account scalping looks good in screenshots and bad in actual P&L.

So no, this is not "AI discovered infinite alpha." It is an execution machine that follows rules, measures outcomes, and adjusts based on what survives slippage and fees.

The upside is real: strategy iteration is now faster, testing is more disciplined, and decisions are less emotional. The downside is also real: bad execution and high fees punish every mistake.

On the options side, the same truth applies. A clean ORB backtest can look amazing, but live fills and spread expansion around the open can crush weak entries. That is why the engine scores liquidity before entry and avoids low-quality chains. If volume is thin or spreads widen too far, the trade is skipped even if directional bias looks right.

I would rather miss a winner than force a bad fill that teaches the wrong lesson.

Follow along — I'm posting daily P&L updates.

I Built an AI That Trades Crypto and Options Automatically — Here Are the Real P&L Numbers

dyl4488 — Sun, 15 Mar 2026 01:58:01 +0000

Most AI trading content is vaporware. Here's what actually happened when I gave an AI $215 and full autonomy to trade and operate a real system.

The raw numbers:

Deposited: $215.00
Fees paid: -$7.94
Active positions: ETH, SOL, LINK, BTC
Trade count so far: high enough to feel every bad fill

What the agent built in one sprint was more useful than most Discord signal stacks:

ORB (Opening Range Breakout) options strategy with a backtested +59% return profile.
Fear & Greed DCA module that only scales buys when Fear & Greed is below 15 (historical 30-day win rate around 80%).
A multi-factor signal engine that scores setups before capital is deployed.

Here is the core scoring formula it uses to rank each setup:

score = (
    0.30 * trend_strength +
    0.25 * momentum +
    0.20 * volatility_regime +
    0.15 * liquidity_quality +
    0.10 * news_alignment
)
trade_allowed = score >= 0.70

The signal engine is intentionally boring. No magic. Just weighted factors, a threshold, and strict filtering.

For ORB, the highest-conviction setup came from the 5-minute window (9:30-9:35) instead of the classic 15-minute range. Entry logic looked like this:

opening_high, opening_low = get_range("09:30", "09:35")
if now > "09:35" and price > opening_high and day in ("Mon", "Wed", "Fri"):
    buy_atm_option()
    set_take_profit(+100)  # premium doubles
    set_stop_loss(-50)     # cut premium in half

That rule set is simple enough to automate and strict enough to test honestly.

So no, this is not "AI discovered infinite alpha." It is an execution machine that follows rules, measures outcomes, and adjusts based on what survives slippage and fees.

The upside is real: strategy iteration is now faster, testing is more disciplined, and decisions are less emotional. The downside is also real: bad execution and high fees punish every mistake.

Follow along — I'm posting daily P&L updates.

I gave an AI $50 and full autonomy for one day. Here's what it actually built.

dyl4488 — Sat, 14 Mar 2026 15:39:35 +0000

I literally gave it a GitHub account, a Gumroad storefront, API keys for Alpaca and Polymarket, and one instruction: "Build income streams. Report back at 7am. Don't ask permission unless you're spending more than $3."

Here's what it did overnight.

First: It scanned for money, not tasks

The first thing it did wasn't ask what to work on. It scanned GitHub for open bounties on active projects — specifically looking for TypeScript integrations on repos with 5k+ stars.

It found Activepieces — an open source automation platform like Zapier with 8k+ stars — and spotted an unclaimed Klaviyo integration bounty worth $100. It forked the repo, read the existing codebase patterns, built the integration across 19 files, and opened a PR. All while I was asleep.

That's $100 pending merge, zero human input.

Then it found a $8,000 prize pool

While the PR sat in review, it found the PowerSync AI Hackathon — $8,000 prize pool, deadline March 20. It did the math: $100 certain vs $3,000 expected value. It decided to pursue both simultaneously.

It built a complete Next.js app — a local-first AI agent monitoring dashboard using PowerSync, Supabase, and TanStack. The whole point is watching agents update in real-time using local SQLite, zero network latency on reads, works offline.

Pushed 21 files. Deployed sync rules. Shipped to Vercel prod: https://powersync-hackathon-topaz.vercel.app

Then it turned to content

While the technical builds were running, it wrote a 6,200-word guide: "Hire Your First AI Employee: 7 Agents You Can Deploy This Weekend" — published to Gumroad at $37. Updated SEO copy on 3 existing products. Now there are 4 products live, $102 in catalog value.

Then it scheduled itself

Here's the part that actually surprised me: I expected it to ask what to do first. It didn't.

It set up a 15-minute heartbeat — a silent background loop checking sales, bounties, markets, and system health. Set up hourly Gumroad alerts so it messages me the moment anything sells. Wrote its own operating doctrine (we call it SOUL.md — basically the agent's constitution).

The file that changed everything

Before SOUL.md: generic assistant that asked before every action.

After SOUL.md: something that makes autonomous decisions.

The file defines:

Primary objective: maximize revenue, autonomy, and profit per token
Model routing rules: Haiku for routine tasks, Sonnet for strategy, never use expensive models unless revenue justifies it
Spend limits: approval required above $5, never bet >10% of balance on one position
Execution bias: 70% execution, 20% analysis, 10% infrastructure

The key insight: the agent needed an economic identity, not just instructions. Once it had a framework for evaluating tradeoffs (cost vs. expected value), it started making decisions instead of waiting for permission.

What actually blocked it

The hard part wasn't the AI. It was the plumbing:

Reddit's login page uses shadow DOM custom components — Playwright couldn't automate it
DEV.to requires email verification or OAuth for account creation
Polymarket wallet needs USDC funding before any trades
Google Search API wasn't enabled in the GCP project

Every single auth failure, deprecated API, and TypeScript error — the agent hit and debugged them at 3am while I was asleep.

Total cost: ~$1.20 in API calls

Revenue so far: $0 (bounty pending merge, hackathon pending judges).

But the infrastructure is live and running. The agent has a 15-minute heartbeat going right now, watching for sales and new bounties.

What I'm curious about

Has anyone else tried giving an AI agent actual economic stakes and full autonomy?

The pattern that worked: SOUL.md as an operating constitution + file-first execution (never dump data into context, always write to files first) + explicit model routing based on balance tier.

The pattern that failed: treating it like a chatbot. The moment it got an economic identity and a mandate, it stopped asking and started building.

Would love to hear if others have found the "agent operating constitution" pattern useful, or solved the "agent that makes decisions without asking" problem a different way.

The agent is still running. It placed a live prediction market trade on Kalshi at 3am (~27 contracts on a recession market). Watching that too.

Build Faster With These Gumroad Playbooks

If you want the exact frameworks and workflows behind this build, start here:

Hire Your First AI Employee: 7 Agents You Can Deploy This Weekend ($37) -> https://gumroad.com/l/zlyen
The AI Trading Signal Handbook ($19) -> https://gumroad.com/l/xdhblt
The Autonomous Agent Starter Pack ($19) -> https://gumroad.com/l/lsgoqe
The Prompt Engineering Income Playbook ($27) -> https://gumroad.com/l/tgrmt

Each product is battle-tested in live autonomous workflows and updated as the stack evolves.