I’ve been active in IT for 15 years, and through this course I’m building my own AI agent. These are my dreams, and the concrete goals that should help me get there. ✨ My dreams - 🏡 Owning my own home. After years of renting, I want something that’s truly mine again. Finding an affordable house in Belgium is a challenge on its own, but I’m not giving up. - 🎁 Giving something back. My parents and my girlfriend have always supported me. My biggest dream is to give something back to them later, simply because I can. - 🎈 Becoming a balloon pilot. Right now I join as crew, but one day I want to stand at the burner myself, with passengers listening to my briefing. - 🌍 Working from anywhere in the world. A laptop, internet, and a good project — that’s all I need. - 🤖 An AI agent that works for me. An agent that learns on its own, acts smartly, and tells me in the evening what it accomplished that day. I come home, eat, and read its report. 😎 🎯 My goals for the next 12 months 1. 💼 A permanent job as a Windows sysadmin. Infrastructure, servers, projects. Remote or on-site with clients, but no more helpdesk. 2. 🤖 A fully functioning trading AI agent. First backtesting, then paper trading, and only going live once the results are solid. With strict safety limits in the code that even my agent can’t override. 😉 3. 📚 Getting back to the theory for my balloon pilot license (BPL). This time with more calm and confidence. 4. 🧠 Learning to truly use AI, so I can apply it in my work as well — for example to monitor infrastructure. 🧭 My motto Haste makes waste. Good things take time, just like a stew that tastes better the next day. 🍲 Are you also working on a trading agent, or dreaming of the skies? Feel free to reach out! 🙌

For my first agent, I’m designing a trading assistant for forex and, eventually, stocks. I’ve mapped the initial file ecosystem in Miro before writing the individual files. The entry point is Agent.md, which defines the agent’s purpose and workflow. It links to Soul.md, which describes my preferences and decision-making style and serves as the central directory for: - Goals: objectives and measures of success. - Rules: shared risk limits and market-specific rules. - Configuration: enabled strategies, instruments and trading periods. - Knowledge: market analysis, strategies, economic news and reference material. - Skills: tasks such as analysing markets, running backtests and evaluating risk. - Tools: connections to trading platforms and data sources. - Memory: observations, trading results and lessons learned. I’m taking an ITIL-inspired approach: clear responsibilities, controlled changes, traceable decisions and continual improvement. The intention is for the agent to propose and test improvements before applying them, while keeping risk limits separate from its learning process. This is an architecture sketch, not a working or proven trading system. I’ll start with historical testing and a demo account, building one part at a time. My biggest takeaway so far: a useful file ecosystem is about giving each piece of information a clear home—not creating as many files as possible. How have you separated knowledge, rules and memory in your own agent projects?

Today I scored six activities based on how much time they take and the value of improving them. The result gave me a clear starting point: a Job Search & Shortlisting Agent. Looking for a junior Windows System Administrator or Windows System Administrator role in East Flanders, Belgium, involves checking several websites, comparing requirements, and deciding which vacancies are worth pursuing. That activity ranked highest in my matrix: - Job searching and shortlisting: 25 - Email review and spam filtering: 16 - Candlestick pattern analysis: 15 My first agent will bring relevant vacancies together in one place, remove duplicates, and explain why each role might suit me. I’ll still decide which jobs to apply for. For me, success means spending less time searching across websites and having a useful shortlist with links, locations, and requirements to work through. I also set a clear personal boundary: checking outstanding payments and making those payments will stay with me. Today’s matrix is saved in agent-priority-matrix.md, alongside my context seed from Day 2. I now have a specific first project to develop—one that could save me time and support my search for a new job.

After the initial dashboard prompt, I made several improvements and extensions: - Redesigned the dashboard with a more modern visual style. - Added a local weather section for Temse above the main header. - Added a login page with a default admin account and mandatory password change on first login. - Added support for multiple user accounts and basic user management. - Added a collapsible left sidebar for navigation. - Added an Account / User Profile section. - Added an MT5 account configuration section. - Added light and dark theme support with saved user preference. - Reworked all dashboard cards so each section has a clear heading inside the box with a separate accent color. - Improved the Stock Watchlist so Yahoo Finance data is more likely to load by using a fallback method when browser CORS restrictions block direct access. - Kept the existing crypto, FX, priorities, weather, news and stock sections integrated into the updated layout.

Day 1 done. Dashboard is now live.