Hi, I'm Maliki Kadafa Tsaqif
Informatics with a product mindset & AI systems focus
I study Informatics at Telkom University in Bandung, specializing in full-stack web development, desktop application engineering, and AI inference deployment. I enjoy bridging the gap between ideas and reality, building robust software solutions using Next.js, React, TypeScript, Go, PostgreSQL, and NVIDIA Triton.
During my Software Engineering Internship at Turkish Aerospace (TUSAŞ) in Ankara, Türkiye, I developed TriGen—a Go desktop application automating NVIDIA Triton Inference Server deployment and ONNX Runtime/TensorRT configurations. Along with projects like the kiosk-style Photobox app and the DashPoint POS system, I bring hands-on experience in AI deployment workflows, local hardware integration, complex database architectures, and clean UI design.
Work Experience

Software Engineering Intern

Key Contributions & Accomplishments
- Developed TriGen, a cross-platform desktop application in Go that automates NVIDIA Triton Inference Server model repository creation and config.pbtxt generation.
- Designed a GUI-based workflow for configuring AI model deployments, reducing repetitive manual setup and improving deployment consistency.
- Implemented automatic generation of Triton-compatible directory structures, model configuration files, and deployment metadata.
- Integrated support for ONNX Runtime and TensorRT deployment configurations with configurable batching, model versioning, and execution settings.
- Built validation logic to detect configuration errors before deployment.
- Researched and implemented deployment workflows using NVIDIA Triton Inference Server, including ONNX export, TensorRT optimization, model repositories, and inference clients.
- Benchmarked CPU and GPU inference performance while evaluating batching strategies to optimize throughput and latency.
Technologies & Tools Used
Featured Projects

TriGen Desktop App
A cross-platform desktop application in Go and Fyne for automating NVIDIA Triton Inference Server deployment. Automatically generates Triton Model Repository structures and validated config.pbtxt files for ONNX Runtime & TensorRT backends with configurable batching, model versioning, and tensor definitions.

DashPoint POS Web App
A full-stack Point of Sale system for local Indonesian stores, featuring role-based access control (Owner, Manager, Cashier) with granular per-employee permission overrides, real-time inventory tracking with low-stock alerts, sales checkout with discount and refund flows, audit logs, and daily sales reports with CSV export.

Photobox Desktop App
A kiosk-style photobooth app for Windows with DSLR and webcam integration, real-time live preview, automated photo compositing into print-ready layouts, silent printing to a dye-sub printer, and instant cloud sharing via QR code.

SD Card Synchronizer
A Java-based utility to synchronize files between SD cards and local storage, plus additional experiments and utilities available on my GitHub.
Technical Skills
I specialize in building high-performance web applications, desktop utilities, and AI inference deployment pipelines. My core stack bridges modern React/Next.js frontend architectures with Go backends and GPU optimization.
Beyond engineering, I care deeply about user experience, accessibility, type safety, and clean software design.
Core Frontend
Backend & Database
AI & Systems Deployment
Styling & UI
Tools & Ecosystem
Let's Work Together
Have a project in mind or just want to say hi? I'm always open to discussing new opportunities and interesting ideas.




