I’ve tried a lot of AI tools with my trading. Most of them were fun to play with for a day and then I’d go right back to doing everything by hand. Grok Bot is the first one I’ve actually been able to rely on with stocks. But here’s the honest truth: the AI by itself wasn’t the big unlock. The big unlock was pairing it with Rocky, the AI coach that lives inside the PXO trading software I use.
I’m a Rockwell Trading referral partner and may be paid if you buy through my link.
So in this post I want to walk you through what my setup actually looks like day to day, show you a real trade from this week (including what I did when it went against me), and be straight with you about where it falls short.
First, the rule that doesn’t change: I make the calls
Before anything else, let me say this clearly. I make every trade decision myself. The AI never places a live trade without my OK. It scans, it flags, it checks the math, it keeps my records clean. But the decision to put money on the line is mine, every single time.
I think that’s the right way to use these tools. AI is a great assistant. It is not a great boss.
What the AI does for me
Here’s the grunt work I’ve handed off:
- Morning scans, every weekday before the open. Before the market opens, the AI runs scans against my written trading plan. Not some generic "best stocks today" list. My rules. That covers the three strategies I run: the Wheel (selling cash-secured puts), Earnings plays outside the expected move, and covered calls.
- Checking every idea against my minimums. Each idea gets checked against my 60% annualized premium minimum and against support levels. If a trade doesn’t pay enough, or the strike isn’t sitting somewhere that makes sense on the chart, it doesn’t make the list.
- Flagging trades. What lands in front of me is a short list of ideas that already passed my rules. I look at them and decide.
- Keeping my trade log current. This one is huge for me. The AI keeps my PXO trade log and journal up to date from my broker fills. No more "I’ll log it tonight" and then forgetting for a week.
- Running a paper-trading desk. When I want to test an idea without risking real money, the AI runs it on a paper-trading desk first. That’s where new ideas earn their way into my real account, or don’t.
How PXO makes it all work
An AI is only as good as what you give it to work with. That’s where PXO comes in. It’s the home base for everything:
- It holds my trading plans. My rules are written down in one place. When the AI runs a scan, it’s checking against something real and specific, not a vibe.
- Rocky scores my plans and scans. Rocky, the AI coach inside PXO, scores both my trading plans and the scans. It’s like having a second set of eyes asking, "Does this actually follow your plan?"
- The logbook tracks everything. Every trade, my P&L, assignments, expirations. All of it in one spot, so I always know where I stand.
Having the plan, the coach, and the logbook in the same place is what turned AI from a toy into something I use every morning.
If you’re curious about the tool I use, I wrote about it here: PowerX Optimizer.
A real example from this week (still open)
Here’s one from Monday, October 12, 2026.
The premarket scan on my live account flagged an Oracle (ORCL) put: the October 16 $134 strike, 4 days to expiration. The bid was $1.22, which works out to 83.08% annualized ($1.22 / $134 x 365 / 4). One example from my own trading. Results vary. That cleared my 60% minimum easily. On top of that, the strike was sitting right on support, near $134.66. So the trade passed on premium and it passed on the chart.
I looked it over, agreed with it, and placed the trade myself in my live account. I got filled at $1.33, which is 90.57% annualized ($1.33 / $134 x 365 / 4). Better than the bid the scan flagged. But that was only the opening fill, not the final result, as you’ll see. This is one example from my own trading, not a promise. Results vary.
That $1.33 fill opened the trade. The AI did the hunting and the math before I even had my coffee. I made the call.
Then the stock moved against me. ORCL dropped below my $134 strike, and by Wednesday, October 14, I had a decision to make.
I’ll be honest: I normally don’t roll puts. This roll was just an opportunity. Sterling, my Grok Bot assistant, suggested it as a good backup because the stock had dropped so much. With the Wheel, I could easily have just taken assignment. That’s how the Wheel works: get assigned the shares, then sell calls on them. This time I simply chose to take the credit on the put instead of getting assigned. My written plan covers this spot too: if a put goes against me, one option is to roll it at the same strike for a net credit before assignment. Every morning, the scan and a daily check look at my open trades and flag any roll options. The AI flags. I decide.
So on October 14, I bought back the October 16 $134 put at $3.42 and sold the October 23 $134 put at $5.22. That’s a $1.80 net credit, same strike, one week further out. Buying back the first put cost more than I sold it for, but the new put paid more than the buyback cost, so the roll itself put a net credit in my pocket.
Here’s where it stands. Counting every leg, I’ve collected $3.13 per share on this trade so far ($1.33 to open, plus the $1.80 net credit from the roll). That makes my break-even buyback $3.13: if I can close the October 23 put for $3.13 or less, the whole trade is at least a wash. Since the roll, ORCL has rebounded to about $134.47 as of Thursday afternoon, October 15, slightly above my $134 strike. If it finishes below $134 Friday, Oct 23, I’ll most likely take assignment and then sell calls on the shares, which is just the next step of the Wheel. But the option to roll again is still there. The trade is still open, and I don’t know yet how it ends.
Wabi-sabi is the Japanese idea of finding beauty in things that are imperfect, unfinished, and always changing. That’s the wabi-sabi part for me: this trade isn’t perfect, but I stick with my plan and manage it.
This is one example from my own trading, not a promise. Results vary. I’m sharing it because it shows the real job of the AI: not just finding the trade, but watching it after I’m in and putting my options in front of me when things don’t go to plan.
The honest part: AI makes mistakes
I’d be lying if I told you this runs perfectly. It doesn’t. A couple of real examples:
- Junk log rows. The AI has made logging mistakes and created junk rows in my log that had to be cleaned up.
- Day-count math disagreements. I’ve had scans disagree with each other on day-count math, which changes the annualized return. When you’re working with short-dated options, a day here or there makes a real difference in that number.
So here’s what I’ve learned:
- Verify. Don’t take the numbers on faith. Check them, especially before you put a trade on.
- Count every leg. After a roll, it’s easy to look only at the newest put. But a "wash" means the whole trade breaks even, so count every leg, not just the newest one. On the ORCL trade above, that’s $1.33 + $1.80 = $3.13 per share collected so far, not just the $1.80 from the roll.
- Keep your rules written down. When the AI and I disagree, the written plan settles it. If your rules only live in your head, the AI has nothing solid to check against, and neither do you.
- Stay the decision maker. The AI is fast and tireless. It is not responsible for your money. You are.
If you want to try this yourself
You don’t need a fancy setup to start. Here’s what I’d tell a friend:
- Write your rules down. What strategies do you trade? What’s your minimum premium? Where do you want your strikes relative to support? Get it on paper. This is the foundation for everything else.
- Start with paper trading. Let the AI run your ideas on paper first. You’ll learn how it thinks, where it gets things wrong, and whether your rules actually hold up, all without risking a dime.
- Let AI do the grunt work. Scanning, math, logging, journaling. That’s the stuff that eats your time and that a machine is good at. Save your energy for the part that matters: the decision.
That’s really it. Cash flow is this easy when you’ve got a plan, good tools, and you stay in the driver’s seat.
If you missed them, start with How I Run the Wheel and How I Trade Earnings.
Want to see where the van was parked while these trades happened? We’re on YouTube at @CJVanVentures.
Talk soon,
Jay
Not financial advice. I’m sharing what I do with my own money, for education only. Options involve risk and are not suitable for everyone – you can lose money, including more than you expect. Read the OCC’s "Characteristics and Risks of Standardized Options" before trading options. I’m a Rockwell Trading referral partner and may be paid if you buy through my link.