Middle LLM Engineer (proficiency in C++)

DeepX is looking for LLM engineer (Middle).
Core Requirements (Must-Haves):

• Mandatory Commercial Experience:

  - If you have a background as a developer (e.g., Python Backend, MLOps, Machine Learning, NLP), you must have at least 1 year of dedicated commercial experience with LLMs.

  - If you started your career directly with LLMs, you must have 1.5-2+ years of commercial experience.
  - Additionally, proficiency in C++ is required.

• Commercial RAG Application: Proven experience implementing RAG (Retrieval-Augmented Generation) solutions in a commercial environment. You should be prepared to discuss in detail which RAG approaches you have used on your projects.
• Deep Technical Knowledge: A fundamental understanding of, and readiness for, a deep technical discussion on topics such as retrieval strategies, data chunking, embeddings, and vector databases.
• Advanced Prompt Engineering: Strong, practical skills in prompt engineering and context engineering, with the ability to discuss your techniques and their impact in detail.

Nice-to-Have:

• Experience working with local or open-source LLMs (e.g., Llama, Mistral).
• Hands-on experience with fine-tuning local LLM models.
• Experience in creating MCP servers.
• Understanding of the A2A (Agent-to-Agent) protocol.
  
  Additional skills:  Qwen, Mistral, Phi, BERT, Transformers, Ollama, vLLM, LoRA,  LLM, NLP, OpenAI, GPT, Anthropic, Claude, Gemini, Gemma, Llama, Mistral, Deepseek,