Aleksa Karac

App Developer & Web Developer

Precision structure, bold creative vision.

About

I build digital products that feel inevitable — where every interaction has purpose and every pixel earns its place. From native apps to immersive web experiences, I obsess over the space where engineering meets craft.

Creative Development

• Three.js / WebGL — Immersive 3D graphics, custom shaders, and interactive web experiences built for performance.
• GSAP — Advanced scroll-driven animations, complex timelines, and buttery-smooth motion design.
• Svelte — Compiled reactive components with zero runtime overhead and elegant state management.
• Interactive Design — Custom cursors, particle systems, physics-based interactions, and generative visuals.
• Motion Design — Scroll-triggered transitions, micro-interactions, and cinematic page reveals.
• SVG Animation — Path morphing, stroke drawing, and vector-based motion graphics for UI delight.
• Responsive Design — Fluid layouts, container queries, and adaptive experiences across every viewport.

Full-Stack Engineering

• React / Next.js — Production-grade web applications with SSR, ISR, and optimized client-side rendering.
• TypeScript — End-to-end type safety across frontend and backend with strict configurations.
• Node.js / Express — RESTful APIs, real-time WebSocket services, and middleware architecture.
• Python — Backend services, automation scripts, data pipelines, and rapid prototyping.
• Rust — Systems-level performance for CLI tools, compute-heavy services, and critical paths.
• Docker / Kubernetes — Containerized deployments, orchestrated scaling, and reproducible environments.
• CI/CD — Automated testing, deployment pipelines, and continuous delivery workflows.
• Linux / Bash — Server administration, shell scripting, and infrastructure management.

AI & Machine Learning

• PyTorch / TensorFlow — Model training, fine-tuning, and inference optimization for production workloads.
• LLM Integration — OpenAI and Anthropic API orchestration, prompt engineering, and agentic workflows.
• Stable Diffusion — Custom image generation pipelines with ComfyUI, ControlNet, and LoRA adapters.
• Video Generation — AI-powered video synthesis pipelines using state-of-the-art diffusion models.
• Fine-Tuning / LoRA — Custom model adaptation with parameter-efficient training for specialized tasks.
• RAG Systems — Retrieval-augmented generation with vector databases and semantic search.
• Hugging Face — Model hub integration, custom inference endpoints, and open-source model deployment.

Let's work together

Get in touch to discuss your project. Email: karacdev@proton.me