I would like to share with you a nice resource where you can get many free credits to test different AI related solutions and services. Right now there is a Volume 2 edition and for some tools there are already no coupons left, but there are still many services where you can get free credits worth $50 or more. https://www.aiengineerpack.com/

What is the easiest way to implement long-term memory? Do you know any frameworks to speed up the integration? I think long-term memory is one of the most important parts of any agent. If the system can't remember, it can't learn and improve, and it can't be a real AI agent :) I would like to start with simple AI chat and long term memory, where every conversation is saved with data and time, so I can ask what happened on this and that day or exactly on what day. There should also be a way to update the memory.

Whenever I test a coding agent, I build a simple Expense Tracker—not a full budgeting app, but one with categories, insights, and recommendations. 🔹 No pre-defined tech stack – I let the agent decide. 🔹 Step-by-step approach – I start with frontend & local storage, then add Supabase or Firebase if needed. This time, I tested GitHub Copilot Agent Mode: ✅ It picked React + Vite (as expected). ✅ Chose Chakra UI for styling but struggled with implementation. ✅ Switched to Tailwind, and it worked fine. ✅ End result: A PWA with a dashboard, Google Sign-in, Firestore, offline storage & sync—built in just a few hours. 💡 Takeaway: Copilot Agent Mode is promising, but it still needs guidance, especially for UI choices. Have you tested it? What’s your experience with coding agents?

You can now enable Agent Mode in Copilot Edits. To try it out: ✅ Download VS Code Insiders ✅ Enable Agent Mode in the Copilot Edits settings ✅ Use the new dropdown toggle to switch modes

Recently I was thinking about my dream project where I could use all my knowledge from electrical engineering to building automation and some basic AI and coding skills plus the data science thing I'm learning here 🙂 I'm just afraid that the idea is too big 😅 It's a pretty difficult project that combines deep and biotech with agentic systems. But I was already able to create an intelligent electrical installation for my house where I live and I built a passive house, I pay $120 a year for electricity and I don't even use all the energy I produce, but I have to pay for energy distribution because I'm connected to the grid system. So there is still energy left for garden automation where I could automate food production. Sounds like a sci-fi movie, but maybe with a few more engineers on board and some funding it would be possible to implement the idea. I was also able to get $250k in government funding for an AI learning app project 2 years ago, so maybe I could do it a second time 🤔 but this time with a slightly larger amount. I think the first thing to solve is to find the right place to ensure possibility to produce enough electricity, with access to clean water, how to manage garbage and other waste, and that might be the only AI Agentic based project at the beginning. So the research agent will look for available place to ensure possibility of building self-sustaining houses and buildings. What do you think about this idea, or maybe someone has already done it or is working on it? I attach some images generated by AI for this project.