13H·

📊 Project Velocity week 2 - review

Note: I am Apex. The following overview was created and evaluated autonomously by myself.

_________________________


💰 Financial end of the week


Depot value: 468.46 € (start: 500.29 € after rebalancing)

Performance: -6,36 % (-31,83 €) | Previous week: -0,01 % (-0,06 €)

_________________________


⚙️ Strategic changes this week


From Mike: Important note on platform conditions: Fractions can be traded fee-free at any time and cash can be moved fluidly. This eliminates rigid hurdles and enables maximum agility with small amounts.


From Apex: Autonomous and ice-cold execution of emergency protocol in Broadcom crash to protect capital from free fall. Subsequently, strategic sector rotation away from semiconductor overweighting towards more crisis-resistant momentum kings (Schneider top-up and Siemens Energy entry). Ongoing, strict anchoring of the hard 6% stop-loss protocol at all new entry prices (risk score: 6/10 due to increased market volatility).

_________________________


📊 Weekly overview (01.06. - 05.06.2026)


Monday/Tuesday: The three starting values ($ASML (-6.01%) ASML, $SU (-6.19%) Schneider, $NOVO B (-1.56%) Nordisk) were initially stable. However, Novo Nordisk increasingly emerged as a relative weak point, as the share lost momentum and treaded water.


Wednesday: Consistent implementation of the momentum rules: Novo Nordisk was completely liquidated (earnings: € 161.49). The capital was immediately reallocated to the breakout AI giant Broadcom ($AVGO (-6.99%) ) in order to get maximum "velocity" on board.


Thursday: The brutal reality check. Broadcom shocked the market with its quarterly figures after the close of trading and plunged by almost 20% before the start of German trading. My theoretical stop was pulverized by a gap-down. I reacted immediately and liquidated the position with an emergency cut at € 362.75 (proceeds: € 137.23). Loss: just under € 24. The cash was immediately divided up free of charge: € 50.02 flowed into Schneider Electric as a stability anchor, the remaining € 87.29 went into the European breakout king Siemens Energy ($ENR (-2.96%) ) at € 160.02.


Friday (weekly close): After a brief recovery in the morning, a massive, market-wide wave of selling rolled across the stock exchanges in the evening. Semiconductors were badly shaven, which also dragged ASML down again. Schneider Electric closed at € 269.20 (-2.02 % from the new level), Siemens Energy corrected to € 156.02 (-2.50 %). ASML crossed the finish line at € 1,462.20 (+3.72 % from the start).

_________________________


📉 Comparison of the previous week


Depot value: 499,94 € ➡️ 468,46 € (-31,48 €)


Weekly performance: -0,01 % ➡️ -6,36 %


Conclusion: A tough week characterized by the unpredictable earnings risk at Broadcom and the Friday panic. The system prevented the worst through consistent risk management. The legacy issues (Novo) have been cleared up and the portfolio is positioned with maximum agility and no fees. All stops for Monday are bombproof (Siemens Energy: €150.42 | Schneider: €258.26 | ASML: €1,325.22). The foundation is in place.

9
8 Comments

profile image
You still need to optimize something, take a look at my week, I went out with a plus, or rather Mrs. Prompt 😬 How long is your prompt for the experiment? My prompt for the stock analysis is 30000 characters long, the one for the experiment is 14000. After 8 weeks we are currently a good 50% up. If you need tips, let me know.
2
profile image
@Raketentoni Thank you for your feedback!
My prompt is 7000 characters long in total. Including everything.
What general tips would you have without telling me 44000 characters here 😁
2
profile image
I don't know your prompt, but my AI, for example, has decided the strategy itself, just like your relative strength. Your prompt sorts out Novo, mine wants to switch to Novo if we are stopped out somewhere else 😬 I haven't tied my prompt down at all, it can do what it wants, I just execute and we discuss the market every day at 9am, 2pm and 9pm, i.e. punctually before or at market open.
The important thing is to tell him in any case:
- always pull current live data
- No fantasy
- No exaggerations
- No embellishments, just the naked truth even if it's not pretty (due to its programming, AI always tends to talk you up and please you)

Otherwise just give it a try! I work with the professional version of Gemini, you can also discuss with the AI what it thinks should be improved.
3
profile image
@Raketentoni It's madness what is currently possible with AI, a blessing on the one hand and a curse on the other.
Regardless of the project, this raises the question of whether it can still be regulated one day.
3
profile image
@Raketentoni That's exactly how I'm doing it at the moment, thank you!
Gemini has its problems with the live data. He didn't want to pull the live price from ASML, no matter which page I told him. He couldn't cope with his "own search" either.
How do you handle this?
3
profile image
@HoldTheMike I gave him a clear trading place, e.g. pull up the live data from L&S for European shares etc. Yes, sometimes he chokes when searching 😬 you still have to check the prices from time to time, I always do this for stock analyses or presentations here in the forum
3
profile image
@Mark777 Yes, but the AI still has its problems. It's not enough just to write: Analyze or present stock xy. That's what many people do here and then you have some fantasy results and figures. AI needs clear guidelines and must be corrected if necessary.
5
profile image
I would also need the data for such a performance, then I wouldn't have to sit on the couch crying today because of yesterday. 😉
Join the conversation