Description

You will lead the end-to-end training and fine-tuning of Large Language Models across open-source and closed-source ecosystems.

Responsibilities

• Architect and implement GraphRAG pipelines using knowledge graphs for contextual retrieval.
• Build and scale distributed training environments utilizing NCCL and Infiniband for multi-GPU setups.
• Apply Reinforcement Learning methods like RLHF, RLAIF, PPO, and DPO to optimize model alignment.
• Innovate using meta-learning, symbolic AI, model fusion, and Mixture of Experts architectures.
• Collaborate with teams to deliver production-grade AI solutions compliant with healthcare regulations.

Required Skills

• Expertise in LLM Training & Fine-Tuning (GPT, LLaMA, Mistral, Qwen).
• Proficiency in Reinforcement Learning (RLHF, RLAIF, PPO, DPO).
• Experience with Graph-based Retrieval Systems, specifically GraphRAG and Knowledge Graphs.
• Familiarity with Distributed Training Frameworks (NCCL, Horovod, DeepSpeed).
• Knowledge of Semantic Search & Vector Databases (FAISS, Weaviate, Milvus).
• Experience with Healthcare compliance and regulatory frameworks (CPT, CM, PCS).
• Background in Symbolic AI & Rule-Based Systems.

Key Skills

Education

PhD or Master’s degree