Krasimir Hristov
Web Developer & AI Engineer
I build intelligent web applications from full-stack products to AI-powered tools and RAG systems. Focusing on precision, performance, and user-centric intelligence.
From Web Dev to AI Engineer
My journey started in the trenches of full-stack development, mastering the art of creating responsive, user-friendly interfaces. As the web evolved, so did my focus — transitioning into the world of Artificial Intelligence.
Today, I specialize in bridging the gap between traditional software engineering and advanced LLM implementation. Whether it's architecting a robust PostgreSQL database or fine-tuning a LangChain retrieval pipeline, I ensure every line of code serves a purpose.
Technical Stack
System Architecture
Scalable RAG and AI pipelines
Specializations
RAG Systems
Vector search, semantic document retrieval and knowledge pipelines. Build Q&A systems that answer from your own data with precision.
AI Chatbots
Context-aware conversational AI with streaming responses, memory management, and custom personas built on modern LLMs.
AI Agents
Autonomous agents with tool-calling, multi-step reasoning, and real-world action execution. From research bots to workflow automation.
Full-Stack Web Apps
End-to-end Next.js applications — from database schema to deployed product. TypeScript, REST APIs, authentication, and responsive UIs.
LLM Integration
Connect any LLM to your product via OpenAI, Anthropic, or OpenRouter. Streaming, function calling, and rate limiting included.
Building with AI Agents
I leverage AI agents, tools, and workflows not as a shortcut, but as a force multiplier. I architect the solutions, define the workflows, and stay in full control — AI handles the execution speed.
Agentic Coding
Cursor, Windsurf, GitHub Copilot, and Cline — each configured with custom API keys and fine-tuned for high-precision output across every stack and project type.
MCP Servers
Custom Model Context Protocol servers that extend agent capabilities with memory, file access, web tools, and domain-specific integrations for richer, context-aware outputs.
Custom AI Workflows
Purpose-built skills, reusable prompt chains, and automation sequences designed to eliminate repetition and compress complex tasks into single commands.
Full-Stack Awareness
I review, understand, and own every line the agent produces. No black-box code — only verified, production-ready output I can explain, debug, and extend at any layer.
Key Differentiator
This is not vibe coding. I use AI the way a surgeon uses precision instruments — with expertise, intention, and complete mastery of the result. AI accelerates execution; I own the architecture.
Projects
SavageAI is an AI chat platform where brutal AI characters a grumpy grandpa, a Balkan dad, a corporate shark actually solve your problems, just without the politeness. Users can upload their own documents and the AI references them during the conversation, giving personalized, context-aware answers while staying in character.
AXON is a personal AI assistant that remembers you. It stores notes, documents, and past conversations, and automatically retrieves relevant context using a RAG system. Supports YouTube summarization, file uploads, and text-to-speech. Currently in active development.
Contact
Initiate Connection
Have a project in mind, a collaboration proposal, or just want to say hello? Open a channel — I'm always available for the right signal.
Direct Channel
Signal Line
+49 173 3079213Node Location
Stuttgart, Germany (Remote-First)