Designed to prevent item evasion and fraud in automated self-checkout systems by cross-referencing visual YOLOv8 object detection against physical barcode scans in real-time.
Computer Engineer Building Autonomous AI Systems & Industrial Vision Solutions.
Computer Engineering (Hons) graduate from UTeM, available immediately. Cisco CCNA Certified & Festo Industrial AI Specialist. Built an autonomous multi-model AI agent (Parliament of Minds) running 24/7 on edge hardware, real-time YOLOv8 anti-evasion vision systems, and network automation solutions. Open to relocation.
Muhammad Irfan Fahmi
- 👤 Full Name Muhammad Irfan Fahmi
- 📍 Base Location Klang Valley, Malaysia (Open to Relocate)
- 🎓 Degree & Merit B. Computer Engineering (Hons) • UTeM
- 🗣️ Languages Malay (Native), English (Fluent)
- 🎖️ Military Reserve Military Reserve (Volunteer) (Resilience)
- 💼 Career Focus Network / AI / Cybersecurity
- 🟢 Availability Available Immediately
Bridging Hardware, Networks & Autonomous AI
I build systems that think. My philosophy: Build → Test → Break → Understand → Improve. From YOLOv8 anti-fraud vision at checkout counters to a Parliament of Minds AI agent making consensus decisions 24/7 on edge hardware — I turn engineering problems into autonomous solutions. Available immediately for full-time roles in AI engineering, networking, or industrial automation.
Featured Projects & Systems
An autonomous AI agent running 24/7 on a dedicated Samsung Galaxy A54 Android node (Termux). Features a specialized Dual-Engine Architecture separating high-frequency automated trade execution from AGI cognitive orchestration, integrated with a Judea Pearl Level 3 Counterfactual Causal Engine that synthesizes invariant operational rules (noise-immune 2.50 pt SL floor, +3.0 pt trailing breakeven) and an active Trading Gate (95+ closed trades, 57.9% win rate, 2.16 profit factor).
An interactive web mini arcade featuring intelligent AI Tic Tac Toe with minimax decision tree logic, soccer penalty shootout mechanics, and sports card memory match. Deployed on Cloudflare edge under custom subdomain with ultra-responsive gameplay.
Zero-touch reading assistant for PC and mobile webtoons to read while eating without touching greasy screens. Powered by proprietary 71D hand geometry feature engineering and a 100-tree Random Forest classifier trained directly on real video demonstrations, achieving 98.6% cross-validation accuracy, zero-recoil geometric classification (eliminating return-stroke false triggers), 0.30s debounced scrolling, and virtual air buttons.
Core Competencies & Tooling Matrix
- Cisco Routing & SwitchingCCNA
- OSPF Protocol & VLANsAdvanced
- Wireshark & Packet CaptureExpert
- Firewall ACLs & SecurityVerified
- Fiber Optic SplicingHands-On
- YOLOv8 Object DetectionPyTorch
- OpenCV Video ProcessingPython
- Festo Industrial AICertified
- Scikit-Learn Machine LearningSVM
- Dataset Augmentation & NMSVerified
- Arduino & ESP32 FirmwareC/C++
- HX711 Strain Gauge SensorCalibration
- USB Serial HID Barcode ScannerIntegration
- EasyEDA Schematic DesignPCB
- Workstation & MaintenanceHardware
- Python ProgrammingCore
- SQL (MySQL & PostgreSQL)Database
- FastAPI Backend ServicesREST API
- Cloudflare Worker & TurnstileEdge Proxy
- Git & GitHub AutomationDevOps