Engineer, LLM

Job Summary


Responsibilities

• Fine-tune foundation models (LoRA/QLoRA, RLHF/DPO, instruction tuning) for domain-specific tasks and terminology.
• Build agentic and tool-use workflows that connect the LLM to internal engineering tools and data sources.
• Design evaluation harnesses for factuality and accuracy; own hallucination and safety guardrails.
• Build RAG pipelines (Azure AI Search) and structured tool-use/function-calling systems on top of fine-tuned Azure OpenAI models.
• Deploy and optimize inference (quantization, KV-cache, vLLM/TensorRT-LLM) on Azure Kubernetes Service for cost-efficient serving.

Qualifications

• 8+ years overall ML/AI engineering experience, with deep hands-on LLM/foundation model work not just calling APIs.
• Demonstrated experience fine-tuning models (LoRA/QLoRA/PEFT, full fine-tuning, or pretraining at some scale).
• Practical experience with alignment techniques (RLHF, RLAIF, DPO, or instruction tuning).
• Strong Python and deep PyTorch proficiency; solid grasp of transformer architecture internals.
• Hands-on experience building RAG pipelines, embeddings/vector search, and agentic or tool-use LLM systems.
• Familiarity with Azure OpenAI / Azure AI Foundry or an equivalent cloud LLM platform.

Preferred Qualifications

• Direct experience with Azure OpenAI Service and Azure AI Foundry for enterprise-scale deployment.
• Experience orchestrating LLM-driven code or structured-output generation for technical/engineering domains.
• Distributed training experience across multi-GPU/multi-node clusters (DeepSpeed, FSDP).
• Publications, blog posts, or open-source contributions in generative AI.

Disclaimer: This job posting has been aggregated from external source. Role details, content, and availability are subject to change. Applicants are advised to confirm the latest information directly on the company website before applying.