Required Qualifications
- 3+ years production ML/AI experience
- Hands-on experience with LLMs beyond prompting
- FastAPI, LLM frameworks (LangChain or similar)
- Experience with embeddings, vector search, RAG
Experience with:
- LLM fine-tuning or distillation (LoRA / SFT / preference tuning)
- Document processing (PDF, OCR), document intelligence, or extraction pipelines
- Designing deterministic or rubric-based evaluation logic
- Working with self-hosted or private LLMs
- Cloud deployment (GCP/OCI) with GPU infrastructure
Preferred Skills (Nice to Have)
- Familiarity with Qwen, LLaMA, DeepSeek, or similar models
- Experience with vector databases and GPU inference
- Experience with enterprise environments
Expected tasks from role:
- Design and implement end-to-end LLM pipelines
- Build information extraction and structured evaluation workflows
- Implement RAG, validation, and confidence mechanisms
- Work with on-prem / private-cloud model deployments
- Optimize AI systems for consistency, reliability, and scalability
