Senior ML Engineer, AI Risk & Compliance Platform

AI Pulse; San Jose (CA); Sur place; USD 187 000 - 438 000; Plein temps

Résumé du poste

TikTok USDS Joint Venture LLC in San Jose is seeking a Senior Machine Learning Engineer to lead the architecture and execution of the AI Risk & Compliance Intelligence Platform and Intelligent Data Agent Ecosystem. You will bridge prompt engineering with native model-driven reasoning and tackle offline/online evaluation, synthetic data generation, and agent skill optimization across DS/DE/SRE workstreams.

You will design and optimize LLM evaluation pipelines, develop ML-driven decision models

Qualifications

• 5+ years of software/ML engineering experience with production ML/LLM systems.
• Deep hands-on expertise with LLM architectures, fine-tuning, RAG, prompt tuning, and agentic frameworks.
• Experience building robust LLM evaluation frameworks and data bootstrap for scarcity/imbalance.

Responsabilités

• Design, build, and scale LLM evaluation architecture and benchmarking pipelines.
• Innovate on synthetic data generation to tackle data scarcity and long-tail risk scenarios.
• Develop ML-driven decision models from complex case data and improve agentic systems.
• Architect, evaluate, and optimize modular agent skills and tools for DS/Analytics/DE/SRE workflows.
• Implement advanced evaluation methods like LLM-as-a-judge and multi-agent benchmarking.
• Provide technical leadership by establishing core ML system design patterns and best practices.

Connaissances

Senior ML Engineer

LLM architectures

LLM evaluation

Data pipelines

Description du poste

TikTok USDS Joint Venture LLC in San Jose is seeking a Senior Machine Learning Engineer to lead the architecture and execution of the AI Risk & Compliance Intelligence Platform and Intelligent Data Agent Ecosystem. You will bridge prompt engineering with native model-driven reasoning and tackle offline/online evaluation, synthetic data generation, and agent skill optimization across DS/DE/SRE workstreams.

You will design and optimize LLM evaluation pipelines, develop ML-driven decision models