Early-Career ML Engineer: LLM & NLP Systems

TikTok USDS Joint Venture; San Jose (CA); On-site; USD 137,000 - 259,000; Full time

Benefits offered by this job

Medical, Dental, Vision Insurance

401(k) with company match

Paid parental leave

Disability coverage

Job summary

TikTok USDS Joint Venture is hiring a PhD-level researcher to architect and build advanced LLM-based systems for processing high-dimensional data across diverse sources. The role emphasizes semantic understanding, including fine-grained intent classification and entity extraction, and applying cutting-edge techniques like RLHF and contextual prompting.

You will optimize model serving for production with quantization and efficient attention, contributing to a robust, scalable AI platform within

Qualifications

• PhD completed or recently completed in Computer Science, Engineering, Math, Statistics, or related field.
• Experience in Machine Learning, NLP, or Computer Vision.

Responsibilities

• Architect and develop sophisticated LLM-based systems to process high-dimensional unstructured data from diverse sources.
• Define and implement complex semantic understanding frameworks, including fine-grained intent classification and entity extraction.
• Apply state-of-the-art techniques to enhance model performance (Instruction Tuning, RLHF/DPO, Contextual Prompt Engineering).
• Optimize model serving for production environments using quantization, distillation, and efficient attention mechanisms (FlashAttention, vLLM).

Skills

NLP

ML

C++

Python

Education

PhD in CS/Engineering/Math

Tools

FlashAttention

Job description

TikTok USDS Joint Venture is hiring a PhD-level researcher to architect and build advanced LLM-based systems for processing high-dimensional data across diverse sources. The role emphasizes semantic understanding, including fine-grained intent classification and entity extraction, and applying cutting-edge techniques like RLHF and contextual prompting.

You will optimize model serving for production with quantization and efficient attention, contributing to a robust, scalable AI platform within