PhD candidate and research engineer working from mathematical signal analysis to deployable sensing systems.
I am a PhD Candidate in Computer Science at Khon Kaen University, working across computational imaging, stochastic signal and image processing, medical AI, and sensor-integrated intelligent systems. My research investigates transform-domain analysis, inverse problems, noise and artifact suppression, and real-time edge inference. In the longer term, I am interested in computational sensory substitution: designing structured mappings between high-dimensional perceptual signals to make information more accessible.
I study signal representations and computational imaging with an emphasis on Fourier and transform-domain analysis, stochastic noise modeling, artifact reduction, inverse problems, and optimization-based reconstruction. My applied work includes medical imaging, ECG and multimodal analysis, AI-assisted detection, and classification, with a focus on methods that can move from controlled experiments toward useful sensing systems.
scikit-image · uv · Virtual Environment · Markdown · IEEE publication workflow · MDPI publication workflow
I design, adapt, and deploy machine-learning systems for computer vision, language, and sensor data. This includes fuzzy logic, YOLO-based detection, LLM and RAG applications, fine-tuning, GPU-accelerated parallel computation, and efficient deployment on NVIDIA Jetson and other edge platforms. I work across experimental training workflows and practical local or containerized inference environments.
YOLO · Fuzzy Logic · LLMs · RAG · Fine-tuning · TensorRT · TensorRT for Jetson · NVIDIA JetPack · Parallel Programming · Pi Agent
I build API-first services and supporting infrastructure for AI applications, real-time systems, and containerized deployments. My backend work spans secure authentication, relational and document data stores, caching, asynchronous messaging, WebSocket communication, reverse proxies, and practical hosting or virtualization environments.
WebSocket · Docker Compose · cPanel
I develop sensor-integrated and real-time systems across microcontrollers, single-board computers, industrial control, and edge accelerators. The work combines hardware–software co-design, signal acquisition, deterministic communication, PCB prototyping, and embedded AI, from early prototypes to operational sensing and control systems.
Ninja · Microcontrollers · Single-board computers · PLC · LoRa · MQTT · CAN · Modbus · RS-485 · Oscilloscope · SiPeed · Hailo · HailoRT · Fritzing · Kivy
I examine the security and resilience of connected systems through authorized research, isolated lab environments, and protocol-level analysis. My interests include network traffic inspection, secure communication, cryptographic primitives, GPU-assisted computation, and the relationship between embedded devices, distributed infrastructure, and cyber-physical risk.
Isolated Lab Environments · TailsOS · SSH Tunnels · tcpdump · hcxdumptool · Hashcat · Threading · Encryption · Decryption · GPU Brute Force · Crypto Analysis
I maintain a small independent infrastructure for private services, remote access, observability, automation, and decentralized-systems experimentation. This space is separate from my security research: it is where I explore practical self-hosting, home automation, personal data services, ad filtering, and Bitcoin or Lightning infrastructure.
RustDesk · Tmux · Herdr · Bitcoin Node · Lightning Node · Crypto Mining · GPU Rig · NerdMiner · 2Miners · NBminer · NiceHash · Binance Future Bot · TradingView API
My work is balanced by gaming, media reviews, podcasts, philosophy, meditation, and knowledge sharing. I also work across several operating systems, so this profile intentionally shows them in order from most frequently used to least frequently used.
Most frequently used → least frequently used · Other Linux distributions are used as needed.
I am currently focused on PhD research, so replies may take longer than usual.
If you are interested in visual perception and image interpretation, explore this project:



















