LLM & RAG Systems
1. Experience with Retrieval-Augmented Generation (RAG)
2. Prompt engineering & context optimization
Chunking, embedding strategies, reranking
Vector Databases
1. Hands-on with Milvus / pgvector / Chroma / Weaviate
Indexing, similarity search optimization
Backend Development
1. Strong in Go / Python / Node.js
2. REST APIs & microservices architecture
AI Model Integration
Experience with:
1. Open-source LLMs (LLaMA, Mistral, etc.)
2. API-based models (OpenAI, etc.)
3. Model orchestration (Ollama, vLLM, etc.)
Data Processing
1. Document parsing (PDF, PPTX, DOCX)
2. Data pipelines for ingestion → embedding → retrieval
Infrastructure
1. Docker / Kubernetes / Docker Swarm
2. GPU-based deployment (AMD / NVIDIA)
3. On-premise AI deployment (important for your case)
Search & Ranking
1. Hybrid search (BM25 + vector)
2. Reranking models (cross-encoder, etc.)
Advanced / Preferred Skills
1. Fine-tuning embedding or LLM models
2. Knowledge graph integration (Graph RAG)
3. Multi-agent AI systems
4. Enterprise system integration (ERP/CRM)
5. Security & data isolation (very important for enterprise AI)
1. 5+ years in backend / system engineering
2. 2+ years in AI / LLM / NLP
3. Experience building production-grade AI systems
4. Familiar with enterprise workflows (ERP is a BIG plus)
5. Experience with on-premise deployments (not just cloud)