Building a Portfolio That Actually Converts. 🚀 I just wrapped up the portfolio build for favs._shot, a photography brand focused on raw emotion and bold, editorial imagery. The challenge? They needed a way to streamline their chaotic booking process and stop losing clients in their DMs. The result is a custom-built, responsive website that prioritizes both aesthetics and lead generation. The dark, minimalist design keeps the focus entirely on the art, while the integrated, frictionless contact system turns casual browsers into booked sessions directly from the site. A perfect blend of form and function.

0

62

Cover image for Manufacturing facility managers were flying

Manufacturing facility managers were flying blind, relying on delayed reports and disconnected spreadsheets to monitor multi-line production status. A modern industrial operation needed a central nervous system. I designed and developed FABRIX AI — a high-performance, real-time command center that aggregates data across active production lines and delivers instant visibility into yield, OEE (Overall Equipment Effectiveness), and maintenance alerts. The Challenge Industrial environments generate massive amounts of data, but it's often siloed or presented in outdated, hard-to-read interfaces. The core challenge was twofold: Data Aggregation: We needed to ingest live statuses, targets, and outputs from multiple concurrent production lines without latency. Cognitive Load: Operators on the factory floor needed a UI that instantly communicated critical alerts (like yield drops or material shortages) without requiring them to dig through menus. The interface had to be deeply legible in low-light industrial environments while looking like a premium, modern SaaS product. he FABRIX AI dashboard transformed raw factory data into an actionable, real-time narrative. Facility managers now have a single pane of glass showing global OEE, total output, and granular line-by-line efficiency. The high-contrast alert system dramatically reduces response times to yield drops and maintenance bottlenecks. The fully responsive, single-page application structure allows the dashboard to be displayed on large factory floor monitors or accessed remotely by shift supervisors without losing visual hierarchy.

1

2

156

Cover image for 📌 Project Title
Multilingual AI Customer

📌 Project Title Multilingual AI Customer Support Bot for Telegram 🎯 Project Overview I developed a fully autonomous, AI-powered customer service Telegram bot designed to handle customer inquiries 24/7. Built for a client in the cryptocurrency space, this bot acts as a first line of defense, automatically resolving common user issues, collecting support tickets, and routing complex problems to human admins. 💡 The Problem The client was overwhelmed with customer support requests on Telegram coming from users all over the world. They were spending hours manually answering repetitive questions and struggling to communicate with non-English speaking users. 🚀 The Solution & Features I built a custom Python-based Telegram bot with the following advanced features: 🤖 Smart AI Matching: Uses PyTorch and a HuggingFace NLP model to actually understand what the customer is asking, rather than relying on strict, dumb keywords. 🌍 Auto-Translation in 7 Languages: Integrates real-time translation. If a user speaks Spanish or Chinese, the bot automatically detects it, talks to them in their native language, but translates the collected ticket into English for the admin! 🔄 Seamless Live-Chat Handoff: If the AI can't solve the issue, it seamlessly connects the user to a human admin for a live 1-on-1 chat directly through the bot. 💻 'One-Click' Auto-Installer: I designed a custom PowerShell installer so the non-technical client could set up and host the bot on their own laptop with a single double-click. 🛠️ Tools & Technologies Used Python (Core logic and flow engine) Telegram Bot API (User interface and messaging) PyTorch & HuggingFace (Machine learning & Natural Language Processing) deep-translator (Real-time multilingual support) 📈 The Impact The bot successfully automated the client's tier-1 support, instantly answering FAQs and organizing complex issues into neat, translated tickets for the admin. This saved the client dozens of hours a week and provided instant, 24/7 support to their global user base.

0

69

Cover image for Built a fully autonomous 24/7

Built a fully autonomous 24/7 automation platform that runs on a remote Linux server with zero human intervention. The system scrapes live data from a heavily obfuscated web application, where CSS class names are randomized on every single page load. I built a custom DOM traversal algorithm using Selenium that bypasses this by using relational element positioning and SVG path analysis instead of standard selectors. On top of the scraper, I built a statistical decision engine backed by SQLite that tracks 10,000+ historical events in real time, calculates live win probabilities, and applies a modified Kelly Criterion formula to dynamically adjust risk parameters per session completely automatically. The platform manages 4 independent headless browser sessions concurrently on a 1GB RAM cloud server, with a custom Flask web dashboard for remote monitoring and configuration. WHAT I BUILT: → Custom Selenium DOM traversal engine (bypasses dynamic obfuscation) → Multi-threaded account manager (4 concurrent headless sessions) → SQLite statistical risk engine with Kelly Criterion staking → Auto-recovery system (handles disconnects, logins, popups) → Flask remote dashboard (live stats, balances, logs, strategy config) → Optimized for 1GB RAM Linux server (systemd, swap, Chromium flags) RESULTS: ✓ 99.9% uptime over extended deployment periods ✓ Sub-400ms execution latency on all critical actions ✓ 10,000+ live events processed and analyzed ✓ Runs for weeks completely without manual input TECH STACK: Python • Selenium • SQLite • Flask • JavaScript • Linux • Bash • systemd • Chromium (Headless)

0

81

Joshua Afolabi

My skill set combines disciplined analytical thinking