Hey everyone! 👋 I have another question for people who are already running their own AI trading systems. I’m currently building my own AI trading project, and I’ve been thinking about something I haven’t really figured out yet: Once the project is completely built and tested, where do you actually run it? Since a trading AI may need to stay active for long periods and continuously receive market data, process signals, monitor positions, and interact with broker APIs, I’m assuming it needs some kind of server or always-on machine. I’d love to know how you guys handle this in real-world projects. For example: - Do you run your trading AI on your personal PC, or do you move it to a VPS/cloud server/dedicated server? - If you use a cloud server, which type of setup do you use? - What kind of CPU, RAM, storage, and GPU does your system actually need? - Does an AI trading system normally need a GPU running 24/7, or can the trading/decision system run mostly on CPU? - How do you handle uptime, internet connection, server restarts, power failures, and system crashes? - Do you run the entire system on one machine, or split different services across multiple machines/servers? - What do you use for monitoring, logs, alerts, and automatic recovery if something goes wrong? - Roughly how much does your infrastructure cost per month? I’m asking because I’m trying to plan the architecture of my project beyond development and testing. I’d especially love to hear from people who have actually taken their trading AI beyond the development stage and are running it continuously in a real environment. I’m still learning, so if I’m misunderstanding anything about how trading systems are normally hosted or operated, please correct me. Thanks! 🙏

Hey everyone! 👋 I’ve been working on a coding project for about a month using MonkeyCode AI, and I’m really enjoying building it. My project has grown quite a lot, and now I want to take it all the way to completion. The problem is that MonkeyCode’s free usage has been reduced from around 30 million tokens to 10 million tokens. Since my project is getting bigger, sometimes even fixing a small bug can use up most or all of my daily tokens, which is making it difficult to continue. So I wanted to ask the community for some advice on two things: 1. Is there a good free alternative? I’m looking for a cloud-based AI coding tool that offers a decent amount of free usage/tokens and can handle larger coding projects, debugging, and code changes. I’ve looked into running open-source models locally, but my PC isn’t powerful enough to comfortably run large coding models. So I’m specifically looking for something cloud-based. 2. Should I pay for MonkeyCode? The other option I’m considering is buying MonkeyCode’s paid plan because I really like my project and want to finish it. I’ve also looked at subscriptions for other coding AI tools, but many of their more powerful plans are around $100+, which is quite expensive for me. MonkeyCode’s most expensive plan is around $60, which seems much more reasonable. However, MonkeyCode isn’t as well-known as some other coding AI tools, and I haven’t found many detailed reviews from people who have actually purchased and used the paid plan. So I’m a little worried about spending $60 and then finding out that the subscription doesn’t provide as much usage as I expected or that there are other limitations. What would you recommend in my situation? - Should I buy MonkeyCode’s ~$60 plan? - Is there a better or cheaper alternative for a large coding project? - If anyone here has actually used MonkeyCode’s paid plan, how much usage/tokens did you get, and what was your experience? I’d really appreciate any advice or recommendations. 🙏

Hey everyone! 👋 I have a question for people who are currently building or have already completed large AI, trading, or software projects. I’m currently building my own AI trading project, and I’m trying to figure out how to manage my time better. I’d like to know: ⏰ How much time do you spend each day? - How many hours do you usually spend planning? - How many hours do you spend actually building/coding? - How much time do you spend on testing, debugging, research, and learning? - Do you work on the project every day, or do you take specific days off? 📝 How do you plan your day? Do you have a specific daily routine, such as: Planning → Research/Learning → Building → Testing → Reviewing → Planning the next day Or do you use a completely different workflow? I’d also like to know how you decide what to work on each day when the project is large and there are many things that need to be done. 💡 Advice for other builders For people who have already completed large projects: How many hours per day do you think is a realistic and sustainable amount of time to spend on a serious project? I’m not asking how many hours someone can work in one day—I’m more interested in what schedule you can realistically maintain for months without burning yourself out. If you could share your own daily routine or time-management method, it would really help me understand how experienced developers approach long-term projects. Thanks! 🙏

Hey everyone! 👋 I have a question for people who have built AI projects, trading systems, or other large software projects. I’m currently building my own AI trading project, and I want to learn from people who have already gone through the process of building and completing a large project. ⏱️ How long did your project take? What is the longest amount of time you have spent building a project from the initial idea/planning stage to a properly working and completed system? Was it: - A few weeks? - A few months? - A year or more? And what made it take that long? 🐛 What problems did you face? I’d especially like to know about problems that you didn’t expect when you started. For example: - Architecture problems - Bugs that were extremely difficult to find - AI-generated code causing unexpected issues - Integration problems between different services - API or data problems - Performance/scaling problems - Database issues - Testing problems - Security problems - Unexpected edge cases - Requirements changing during development - Problems that only appeared when the system became large - Problems that appeared after the project seemed “finished” 🔧 How did you solve them? For each major problem, I’d love to know: What was the problem → how did you diagnose it → how did you solve it → where did you find the solution? For example, did you find the answer through: - Documentation - GitHub - YouTube - Books - Research papers - Online communities - Other developers - AI coding assistants - Trial and error 💡 What should beginners know beforehand? This is the part I’m most interested in. If you could go back to day one of your project, what problems would you warn yourself about? Are there any mistakes or difficulties that you now know are almost unavoidable when building a large project? I’m asking because I’m currently in the middle of building my own project. If I know about common problems before they happen, I can either design around them from the beginning or at least be prepared when they appear.

Hey everyone! 👋 I have a question for people who are building or have already built AI trading systems. I’m trying to understand how you actually communicate complex requirements to an AI when developing a trading system. 🤖 Prompting When you give an AI a prompt to build, analyze, or modify part of your trading system: - How detailed are your prompts? - How do you structure them? - Do you break a large task into multiple steps/phases? - Do you give the AI the complete system context first, or explain only the specific part it needs to work on? - How do you prevent the AI from misunderstanding requirements or making changes that break other parts of the system? - What has been the most difficult part of prompting AI while building your project? - Do you have a specific prompting workflow that works well for large coding projects? I’d especially like to know whether experienced developers use a structured prompt/specification → implementation → testing → review → correction process. 🧮 Mathematics in Trading AI I’m also trying to understand how much mathematics you actually use inside a serious trading AI. For example: - Probability - Statistics - Linear algebra - Calculus - Time-series analysis - Optimization - Regression - Correlation - Volatility - Risk/reward calculations - Position sizing - Portfolio mathematics If you use mathematical formulas, I’d love to know: Why did you choose that particular formula? What exactly do you expect the AI/system to understand or calculate from it? For example, is the formula being used to: - Generate a feature/input? - Measure risk? - Estimate probability? - Detect market patterns? - Calculate volatility? - Score a trading opportunity? - Optimize a decision? - Validate another signal? I’m interested in the reasoning behind the mathematics, not just a list of formulas. 🧠 ML / Data Science / Deep Learning Another thing I’m confused about is how people decide which Machine Learning, Data Science, and Deep Learning techniques to use in a trading AI.