Igor Joshevski

Igor Joshevski

Building with JavaScript & AI

I work with modern JavaScript ecosystems and AI implementation—particularly RAG systems and practical AI integration. My focus is on creating scalable architectures and helping teams navigate complex technical decisions.

JavaScript React & Angular Node.js & NestJS Docker AI & RAG

What I Work With

JS

JavaScript Ecosystems

I've spent years working across the full JavaScript stack—from building interactive frontends with React and Angular to architecting Node.js and NestJS backends. TypeScript has become my go-to for larger applications where type safety matters, and Docker is standard for deployment consistency.

React, Next.js, Angular ecosystems
Node.js, NestJS & distributed systems
TypeScript, Docker, containerization
AI

AI & RAG Systems

Recently, I've been deep in AI implementation—specifically RAG (Retrieval-Augmented Generation) systems that combine traditional data retrieval with LLMs like Llama. I work with voice AI technologies including TTS (Text-to-Speech) and STT (Speech-to-Text) to build practical, reliable applications.

RAG architecture & vector search
Llama, LLM integration patterns
Voice AI: TTS, STT, speech systems

How I Think About Technology

Philosophy and approach to software development and AI system design.

I tend to approach technical problems with a balance of pragmatism and curiosity. Having worked across startups and larger organizations—including high-traffic platforms in media and entertainment—I've learned that the best solutions often come from understanding the context deeply, especially when real-time data and user experience are critical.

When it comes to AI, I'm particularly interested in the gap between what's technically possible and what's practically useful. RAG systems fascinate me because they bridge that gap—taking the power of large language models like Llama and grounding them in real, reliable data. Voice AI (TTS/STT) adds another dimension to how users interact with these systems.

I enjoy helping teams navigate complexity. Sometimes that means containerizing services with Docker, architecting a new NestJS backend, or untangling existing React/Angular code to make it more maintainable. The common thread is always building something that actually works well for the people using it.

Areas of Technical Focus

Long-term interests and ongoing explorations in software development and artificial intelligence systems.

JavaScript Architecture

Scalable Frontend Systems

Designing component architectures that remain maintainable as applications grow. Focus areas include state management patterns, performance optimization, and developer experience across React and Angular ecosystems.

Experience building interfaces for high-engagement platforms where performance and real-time updates are essential.

Keywords: React, Angular, Component Design, State Management, Performance
AI Integration

Retrieval-Augmented Generation

Building systems that combine structured data retrieval with generative AI capabilities using models like Llama. Emphasis on accuracy, reliability, and practical implementation patterns.

Exploring vector search, embedding strategies, and context management for production applications.

Keywords: RAG, Llama, Vector Search, LLM Integration, Embeddings
System Design

Distributed Backend Architecture

Architecting Node.js services and distributed systems that handle scale gracefully. Focus on API design, data flow, and infrastructure decisions.

Experience with microservices, serverless architectures, and the trade-offs involved in each approach.

Keywords: Node.js, Microservices, API Design, Distributed Systems
Type Systems

TypeScript at Scale

Applying type systems to large JavaScript codebases to improve reliability and developer velocity. Balancing strictness with pragmatism.

Interested in gradual adoption strategies and type-driven development practices.

Keywords: TypeScript, Type Safety, Code Quality, Migration Strategies

Let's Connect

Always interested in chatting about JavaScript architecture, AI systems, or interesting technical challenges. Best way to reach me is LinkedIn.

Connect on LinkedIn