We're a 20-person tech consultancy building AI-powered automation solutions for international clients. Our team ships production AI systems: intelligent chatbots, RAG-based document processing, automated workflows with LLM integration, and custom AI solutions for diverse industries. We handle both client consultancy work and develop our own internal products.

What You'll Actually Build

• Production-ready RAG systems using vector databases (Pinecone, Chroma, Weaviate)
• Complex LLM integrations with OpenAI, Anthropic Claude, and other models
• AI automation pipelines using LangChain, LlamaIndex, or custom frameworks
• Intelligent chatbots and conversational AI systems
• Document processing and analysis systems with NLP
• Custom AI workflows integrated with tools like n8n, Make, Zapier
• Fine-tuning and prompt optimization for specific use cases
• Deploy and scale AI solutions in production environments

Tech Stack

Core Requirements:

• Python (FastAPI, Flask)
• LLM APIs (OpenAI, Anthropic, Gemini, etc.)
• LangChain or LlamaIndex (production experience)
• Vector databases (Pinecone, Chroma, Weaviate, or similar)
• RAG architectures (you've built these, not just read about them)
• Prompt engineering and optimization
• API development and integration
• Git, Docker, deployment workflows

Additional Experience:

• Working with embeddings and semantic search
• Function calling / tool use with LLMs
• Streaming responses and token optimization
• Handling context windows and chunking strategies
• AI agent frameworks (CrewAI, AutoGPT, or custom)

What We're Looking For

• 2+ years of Python development
• At least 1 year building production AI/LLM systems (not just tutorials or side projects)
• You've deployed RAG systems that real users query daily
• Experience with vector databases in production
• You understand the nuances: chunking strategies, retrieval optimization, hallucination mitigation
• Can architect AI solutions from scratch, not just follow docs
• Comfortable working with international clients and understanding their requirements
• Strong problem-solving - you debug complex AI behavior independently
• Excellent English communication (you'll talk to clients)

Strong Candidates Have:

• You've built a RAG system that actually works well (and can explain why)
• You know when to use different embedding models and can justify it
• You've dealt with LLM rate limits, costs, and optimization in production
• You can explain the tradeoffs between different vector search approaches
• You've debugged why an LLM is hallucinating and fixed it
• You get excited about prompt engineering techniques and token efficiency

What We Offer

• Fully remote
• Work on diverse, challenging AI projects across multiple industries
• Direct client interaction and ownership of technical solutions
• Both consultancy projects and internal product development
• Small, focused team - your work matters from day one
• Opportunity to shape our AI engineering practices

What You Won't Be Doing

• Training models from scratch (we use APIs and fine-tuning)
• Deep learning research or academic work
• Front-end development (we have a separate team)