[{"content":"","date":"20 April 2026","externalUrl":null,"permalink":"/badges/ai/","section":"Badges","summary":"","title":"AI","type":"badges"},{"content":"","date":"20 April 2026","externalUrl":null,"permalink":"/badges/","section":"Badges","summary":"","title":"Badges","type":"badges"},{"content":"I\u0026rsquo;m currently building a summer house at the seaside. If you\u0026rsquo;ve ever tackled a full-scale construction project, you know it\u0026rsquo;s not simply about picking out tile colors. It’s an exercise in extreme project management: chasing architects, chasing formal authority approvals, navigating local community consultations, wrangling mortgages, and negotiating with a half-dozen vendors simultaneously.\nVery quickly, the scale of the project started growing out of control. With so many dependencies and parallel activities happening at once, I felt like I was losing my grip on the timeline. Given my background in project management, I knew exactly what was happening and what was needed: a proper system to manage the chaos.\nInstead of just installing another app, I decided to build a custom AI Assistant to act as our personal project manager.\nThe Architecture: Tying the Stack Together # The goal was to build a system that had access to all our documentation, understood construction data, and tracked progress. Most importantly, the interface had to be dead simple to use while standing on a muddy plot of land holding a phone in one hand.\nHere is how the stack works:\nThe Frontend (Telegram): I created a Telegram bot named \u0026lsquo;AI Builder\u0026rsquo;. Telegram is ideal for this because it naturally handles text, voice memos, and file uploads directly from a phone. The Brain (n8n \u0026amp; Gemini): Whenever a message is sent to the bot, it triggers an n8n workflow. If the input is voice, n8n passes it to Google Gemini to transcribe the audio into a clean text prompt. The Agent: Once transcribed, the prompt is passed to an AI Agent within n8n. This agent is wired up to a local database, my Google Drive (where the heavy documentation lives), and Jira. Yes, I use Jira for a home construction project—and it works. The Output: The agent executes the necessary operations and fires a confirmation response back to us in the Telegram chat. A Note on Security # Because Telegram bots are essentially public if someone finds the handle, Step One of the n8n logic is a strict security gate. The workflow immediately drops any traffic that doesn\u0026rsquo;t originate from my or my wife\u0026rsquo;s authorized Telegram IDs.\nReal-World Commands # Having this system means I don\u0026rsquo;t have to open a laptop to manage the project. From the plot, I can send voice commands to the bot like:\n\u0026ldquo;Review the house\u0026rsquo;s technical documentation and add a comment to the roof cover task detailing exactly which types of materials we are approved to use.\u0026rdquo; \u0026ldquo;Add a task for me called \u0026lsquo;Review contract with electrician\u0026rsquo; and add a comment: \u0026lsquo;make sure it covers the PV installation\u0026rsquo;.\u0026rdquo; (Uploading a batch of photos) \u0026ldquo;Upload these pictures to the \u0026lsquo;ground leveling\u0026rsquo; task.\u0026rdquo; \u0026ldquo;I have a plumber coming next month. Which tasks do we need to complete before he arrives? Once he is done, what other tasks will we be able to start\u0026rdquo; The agent retrieves the right Jira ticket, pulls context from Drive if necessary, and handles the data entry.\nMultiplayer Mode: A Shared Brain # The biggest advantage is that the \u0026lsquo;AI Builder\u0026rsquo; bot is a shared group chat between me and my wife.\nThis creates a single, visible source of truth. We can both add new tasks, drop in comments, save documents, or log budget updates in one place. We can assign tasks to each other seamlessly. Because it all syncs back to Jira, nothing gets lost in a scrolling chat history. It completely removed the \u0026ldquo;Wait, I thought you called the contractor?\u0026rdquo; conversations.\nThe Technical Roadblocks # Building this wasn\u0026rsquo;t instantly smooth. A few specific headaches stood out:\nWiring the Agents: I chose to manually connect all the necessary tools to the n8n agents myself. Call me paranoid, but I still don’t entirely trust external \u0026lsquo;agent-in-a-box\u0026rsquo; solutions. I prefer the granular control of building the workflow myself, even if it takes longer to set up.\nFighting Atlassian’s API: Fixing the bugs between n8n and Jira swallowed a lot of time. The Jira API makes things unnecessarily complicated because the calling schema changes depending on the specific type of update you are trying to push. It is something different to update a data attribute of an issue and a description of an issue \u0026hellip; why ? \u0026hellip;\nThe PDF Problem: The extraction tool I’m using for the RAG system gives decent results for standard text contracts and permits. However, it completely fails at reading and interpreting architectural technical drawings. That\u0026rsquo;s a bottleneck I\u0026rsquo;m still trying to figure out.\nBuilding your own AI project manager to build a house might sound like overkill, but the time saved on data entry and the mental clarity of having a single point of tracking has a great chance to pay off. We\u0026rsquo;ll see. I am only at the beggining of that journey.\n","date":"20 April 2026","externalUrl":null,"permalink":"/projects/summerhouse/","section":"Projects","summary":"I\u0026rsquo;m currently building a summer house at the seaside.","title":"Bob the (AI) Builder","type":"projects"},{"content":"","date":"20 April 2026","externalUrl":null,"permalink":"/badges/jira/","section":"Badges","summary":"","title":"Jira","type":"badges"},{"content":"","date":"20 April 2026","externalUrl":null,"permalink":"/badges/n8n/","section":"Badges","summary":"","title":"N8n","type":"badges"},{"content":"","date":"20 April 2026","externalUrl":null,"permalink":"/projects/","section":"Projects","summary":"","title":"Projects","type":"projects"},{"content":"","date":"20 April 2026","externalUrl":null,"permalink":"/badges/telegram/","section":"Badges","summary":"","title":"Telegram","type":"badges"},{"content":"","date":"6 January 2026","externalUrl":null,"permalink":"/badges/agents/","section":"Badges","summary":"","title":"Agents","type":"badges"},{"content":"","date":"6 January 2026","externalUrl":null,"permalink":"/badges/api/","section":"Badges","summary":"","title":"API","type":"badges"},{"content":"I’ve spent the last week seeing exactly how much heavy lifting I could offload to an AI.\nThe goal was simple: Build a Chrome extension called noBait to nuke clickbait and distracting thumbnails from my browser. But the real goal was to stress-test the new GLM 4.7 model using OpenCode to see if it could actually handle a project end-to-end without me having to babysit every line of syntax.\nTooling \u0026amp; Setup # I didn’t want to just copy-paste snippets from a chat window. I wanted an \u0026ldquo;agentic\u0026rdquo; workflow.\nEarly on, I bounced around between a few different AI CLI tools, trying to find the right \u0026ldquo;vibe.\u0026rdquo; After a few failed starts, I decided to stop window shopping and just master one tool: OpenCode. It became my primary interface for everything.\nThe split between \u0026lsquo;Plan\u0026rsquo; and \u0026lsquo;Build\u0026rsquo; agents seem to work perfectly well. Usage of Agent.md acted as the project\u0026rsquo;s \u0026ldquo;brain\u0026rdquo;—defining the architecture and coding standards. By feeding this into OpenCode, the model (GLM 4.7) actually had a map of the territory. It wasn\u0026rsquo;t just guessing; it was working within the constraints of my existing file structure.\nTesting GLM 4.7 Limits # Most people use these models for simple scripts. I have a feeling this try was a little bit more demanding, where this new model had a chance to shine.\nWhat worked surprisingly well:\nContext Retention: Because of the Agent.md setup, I didn\u0026rsquo;t have to keep reminding the model about my specific selectors. It \u0026ldquo;remembered\u0026rdquo; the project\u0026rsquo;s soul. And it general it managed this really well, even when comparing to other models. Logic over Syntax: GLM 4.7 is sharp. When I asked it to handle various websites loading mechanisms, it correctly suggested a MutationObserver instead of a lazy setTimeout loop. That said, it wasn\u0026rsquo;t a perfect solo run. There were a few moments where GLM got stuck in a loop and I had to pivot. In those cases, I brought in Grok and Gemini to provide a second opinion or break a logic stalemate. Having that \u0026ldquo;council of models\u0026rdquo; was key to getting past the tricky parts.\nWhy I’m Calling it \u0026ldquo;Done\u0026rdquo; # The project is about 99% there. If you check the repo, the extension works. It cleans up the UI, the popup is functional, and the core logic is solid.\nIt still needs some polish for cross-browser usage and better multi-language support, but I’m marking it as complete for a very specific reason. During my final push, I discovered the \u0026lsquo;Oh my OpenCode\u0026rsquo; orchestration add-on.\nSeeing what that orchestrator can do made me realize that noBait has served its purpose as a learning ground. Between the free tier limits on my OpenRouter API and this new discovery, I’ve lost interest in tweaking secondary functions for cross-browser support. I see a much more challenging and interesting project on the horizon that actually demands that kind of advanced orchestration.\nnoBait proved that with a well-structured Agent.md and the right focus on your tooling, you can move incredibly fast. Now, it\u0026rsquo;s time to take what I learned and apply it to something even bigger.\nRepository: vermibor/noBait\nTooling: OpenCode + GLM 4.7 (with assists from Grok \u0026amp; Gemini)\n","date":"6 January 2026","externalUrl":null,"permalink":"/projects/nobait/","section":"Projects","summary":"I’ve spent the last week seeing exactly how much heavy lifting I could offload to an AI.","title":"The noBait Experiment","type":"projects"},{"content":"","date":"6 January 2026","externalUrl":null,"permalink":"/badges/weekend-project/","section":"Badges","summary":"","title":"Weekend Project","type":"badges"},{"content":"So here\u0026rsquo;s a problem I never thought I\u0026rsquo;d automate away: figuring out what to do on weekends.\nLiving in a small village has its perks – quiet streets, friendly neighbors, that whole peaceful countryside vibe. But when Friday evening rolls around and you realize you have zero plans? Gardening is always an option but maybe there is something more interesting. My wife and I would end up in this Thursday/Friday night ritual where I\u0026rsquo;d spend an hour or two hunting through Facebook events, local forums, and random websites trying to find something interesting happening within reasonable driving distance.\nThe local social media groups? They\u0026rsquo;re mostly people selling furniture and complaining about potholes. Not exactly a goldmine for weekend inspiration.\nEnter: The Weekend Concierge # Last weekend I finally had enough time and decided to build something about it. I fired up n8n (my favorite automation tool these days) and spent a few hours putting together what I\u0026rsquo;m calling my \u0026ldquo;AI concierge\u0026rdquo; agent.\nHere\u0026rsquo;s what it does: every week, it automatically searches for events, activities, and interesting stuff happening in our area. Then it uses AI to filter through everything based on what my wife and I actually like – no random concerts, networking events or kids\u0026rsquo; birthday party venues, thank you very much. Finally, it packages everything into a neat little newsletter that lands in our inbox. And it learns to be better each time.\nHow It Works # The workflow isn\u0026rsquo;t rocket science, (it is a simple recommendation engine after all using RAG architecture) but it feels like magic when it works:\nData gathering – It pulls events from various sources (local websites, event aggregators, even some scraped content from regional tourism pages) AI filtering – This is what makes the difference. I feed all the raw event data into an AI model with a prompt about our preferences: outdoor activities, live music, food markets, anything cultural but not too fancy, that kind of thing. And it is also linked with a database of past events we liked the most. Newsletter generation – The filtered results get formatted into a readable email with dates, locations, and why the AI thinks we might like each thing Weekly delivery – Every Wednesday morning, there it is in our inbox The Best Part # I don\u0026rsquo;t think about it anymore. That\u0026rsquo;s the whole point. No more \u0026ldquo;I guess we\u0026rsquo;ll just stay home again\u0026rdquo; weekends because I forgot to plan ahead.\nAnd honestly? The AI does a better job than I did manually. It catches stuff I would\u0026rsquo;ve missed, and it\u0026rsquo;s way better at connecting the dots between what we liked before and what might interest us next.\nWorth the Time? # Took me maybe 3-4 hours to set up including all the testing and tweaking. I\u0026rsquo;ve already saved more time than that just in the past month, and we\u0026rsquo;ve actually done more interesting things because the planning friction is gone.\nBut hold on \u0026hellip;. can I make it as a personal, super focused event recommendation service to wider public that will look at individual preferences? I may spend more time on enhancing this simple flow. In the meantime: download this n8n workflow and set this up for yourself. Just add your credentials and change preferences in \u0026lsquo;parameters\u0026rsquo; step. Enjoy.\n","date":"7 December 2025","externalUrl":null,"permalink":"/projects/ai-concierge/","section":"Projects","summary":"So here\u0026rsquo;s a problem I never thought I\u0026rsquo;d automate away: figuring out what to do on weekends.","title":"AI Concierge","type":"projects"},{"content":"","date":"7 December 2025","externalUrl":null,"permalink":"/badges/automation/","section":"Badges","summary":"","title":"Automation","type":"badges"},{"content":"","date":"25 May 2025","externalUrl":null,"permalink":"/badges/crypto/","section":"Badges","summary":"","title":"Crypto","type":"badges"},{"content":"","date":"25 May 2025","externalUrl":null,"permalink":"/badges/simple/","section":"Badges","summary":"","title":"Simple","type":"badges"},{"content":" Building a Crypto Correlation Explorer with Vibe Coding # How it started # I wanted to visualize cryptocurrency relationships, so I spent an afternoon vibe coding to see how far I could get. I\u0026rsquo;ve been interested in how different tokens correlate and move together, and building an interactive explorer seemed like a good way to figure it out.\nThe Project: Crypto Correlation Visualization # What I Built # Using Google AI Studio, I crafted an application that:\nVisualizes token relationships through interactive correlation matrices Analyzes log returns to understand price movement patterns Explores statistical relationships between different cryptocurrencies Backtests a trading strategy which uses those relationships Do you want to see results? Check it out! Technical Details # The statistical foundation includes:\nLog return analysis for normalized price comparisons Correlation mapping to identify strongly related token pairs Cointegration tests to validate if two pairs move together (Johansen\u0026rsquo;s test) These statistical tools are crucial for understanding whether crypto pairs represent a true relationship - essential knowledge for any serious trading strategy.\nFuture Plans # This isn\u0026rsquo;t just another data visualization project. I\u0026rsquo;m building toward something bigger: a full-scale trading platform focused on market-neutral strategies like pair trading.\nImagine identifying crypto pairs that historically move together, then capitalizing on temporary divergences through sophisticated algorithmic trading. That\u0026rsquo;s the endgame.\nWhy vibe coding worked here # For this project, vibe coding meant skipping heavy upfront planning and just following the data. I started with simple visualizations and let the tool evolve naturally as I noticed interesting correlation patterns.\nWhat\u0026rsquo;s Next? # The roadmap is exciting:\nImplement cointegration testing algorithms 🏁 Add automatic data refresh 🏁 Build backtesting capabilities 🏁 Integrate trading strategy with your Ex/Dex of choice Quick update: # After spending one more evening on my application it now has a strategy builder module. You can test how a market-neutral strategy performs depending on various parameters. You can also use AI-driven helper! I still have a few issues to solve: speed is one of those. When you click \u0026lsquo;Simulate\u0026rsquo; - give yourself a break, make a coffee, call a friend or watch next episode of your favourite show. It\u0026rsquo;s been published under 3 domains:\ncryptopair.xyz\ncryptopairs.xyz\ncryptocorrelations.xyz\n","date":"25 May 2025","externalUrl":null,"permalink":"/projects/vibecoding/","section":"Projects","summary":"Building a Crypto Correlation Explorer with Vibe Coding # How it started # I wanted to visualize cryptocurrency relationships, so I spent an afternoon vibe coding to see how far I could get.","title":"Vibecoding","type":"projects"},{"content":"","date":"25 May 2025","externalUrl":null,"permalink":"/badges/diy/","section":"Badges","summary":"","title":"DIY","type":"badges"},{"content":"","date":"25 May 2025","externalUrl":null,"permalink":"/badges/iot/","section":"Badges","summary":"","title":"IoT","type":"badges"},{"content":"","date":"25 May 2025","externalUrl":null,"permalink":"/badges/rpi/","section":"Badges","summary":"","title":"RPi","type":"badges"},{"content":"What can I do with my unused Raspberry Pi that previously worked as retro game console, NAS server and Kodi media center? Well \u0026hellip; My house has several different sensors: movement detectors, door opening switches, power and water meters and so on. Let\u0026rsquo;s add some more sensors for weather, heating system, number of people around and then I can \u0026hellip; For now this project is only about collecting data. I may use it some day.\n","date":"25 May 2025","externalUrl":null,"permalink":"/projects/smarthome/","section":"Projects","summary":"What can I do with my unused Raspberry Pi that previously worked as retro game console, NAS server and Kodi media center?","title":"Smarthome","type":"projects"},{"content":" How I Built an Automated Workflow to Hunt Trends \u0026amp; Launch Niche Websites # Staying ahead of trends and picking the right one to pursue usually takes a lot of manual research. To fix this, I built an automated system that handles the heavy lifting for me.\nWhat I Created # Using advanced workflows in n8n, I set up an automated pipeline that:\nScans newsletters and data sources to identify trending topics Applies quality filters to select the most promising trends Randomly picks a weekly set of trends to analyze further Crafts a strategy for simple, niche websites based on these trends Uses AI workflow to estimate the value and potential of each site Passes the best candidates to me - the human in the loop - for final approval Automatically executes website creation and deployment through my hosting provider This automation lets me launch new niche sites lightning-fast, freeing up time to focus on growth and strategy.\nNext steps: Optimize SEO and content management to boost traffic and engagement.\nWhy I’m Doing This # Learn by doing: Each project helps me sharpen my automation and domain-building skills Build a portfolio: Over time, I’ll create a collection of valuable domains that could be auctioned or monetized Test cutting-edge AI \u0026amp; automation tools in a real-world workflow Tools \u0026amp; Technologies Powering the Project # Category Tools \u0026amp; Services Workflow Automation n8n Project Management Jira Alerts \u0026amp; Monitoring Pushover Cloud Hosting Google Cloud (self-hosted n8n) Communication \u0026amp; Storage Google APIs (Gmail, Drive) Database Search Pinecone Domain \u0026amp; Hosting Setup NameCheap + cPanel APIs AI Models Used # Gemini 2.5 ChatGPT 4.0 mini and 4.1 mini DeepSeek R1 Claude Sonnet 3.5 I use a mix of models depending on the task—some are better at structuring data, while others handle summarization or strategic weighting.\nProject Performance Audit # Visit this page again soon - the most interesting stuff will be uploaded here.\nUpdate # Project has turned into something much bigger. Or at least some core part of this. Check out soon for details.\nWant to know more about how this works or get insights on automating your own projects? Feel free to reach out or follow along as I refine this process!\n","date":"25 May 2025","externalUrl":null,"permalink":"/projects/domains/","section":"Projects","summary":"How I Built an Automated Workflow to Hunt Trends \u0026amp; Launch Niche Websites # Staying ahead of trends and picking the right one to pursue usually takes a lot of manual research.","title":"AI Domain Advisor","type":"projects"},{"content":"Hello! I\u0026rsquo;m a hobbyist exploring the fascinating world of Artificial Intelligence and DIY tech during my afterhours. My journey started with the fundamentals of machine learning, and I\u0026rsquo;ve since delved into deep learning, computer vision, and natural language processing. I find immense satisfaction in bridging the gap between theoretical concepts and tangible, working projects.\nDriven by curiosity, I continuously experiment with new algorithms, frameworks, automation platforms and hardware. This showcase is a collection of some of my explorations, ranging from practical tools to purely experimental concepts.\nBeyond the code, I believe in the potential of AI to be a creative and problem-solving tool. My focus is often on building systems that are not only functional but also intuitive and perhaps even a little fun. I\u0026rsquo;m always learning and eager to tackle the next challenge.\nSee my profile of a standard corpo life on LinkedIn. ","date":"24 May 2025","externalUrl":null,"permalink":"/about/","section":"Homepage","summary":"Hello! I\u0026rsquo;m a hobbyist exploring the fascinating world of Artificial Intelligence and DIY tech during my afterhours.","title":"Something about me","type":"page"},{"content":"Hello Visitor! Thanks for stopping by. Explore some of my recent project, and if you would like to know what will come next: sign up for Newsletter\n","date":"24 May 2025","externalUrl":null,"permalink":"/","section":"Homepage","summary":"Hello Visitor!","title":"Homepage","type":"page"},{"content":"","externalUrl":null,"permalink":"/authors/","section":"Authors","summary":"","title":"Authors","type":"authors"},{"content":"","externalUrl":null,"permalink":"/categories/","section":"Categories","summary":"","title":"Categories","type":"categories"},{"content":"","externalUrl":null,"permalink":"/series/","section":"Series","summary":"","title":"Series","type":"series"},{"content":"","externalUrl":null,"permalink":"/tags/","section":"Tags","summary":"","title":"Tags","type":"tags"}]