Contribute to unified AI/ML capabilities by integrating classical ML models with Generative AI platforms (primarily AWS Bedrock) for applications in semiconductor manufacturing and risk analytics.
Design, build and operate production-grade GenAI systems in Python: agents and tool use (MCP, function calling), RAG backends, document parsing pipelines, APIs and containerised services on Azure.
Deep Learning Model Implementation & Product Development Ownership: Build, train, evaluate, and maintain CNN/U-Net/ViT models for automated inspection and measurement. Own model performance end-to-end — ablation studies, confidence calibration, product development performance monitoring.
Design, build and operate production-grade GenAI systems in Python: agents and tool use (MCP, function calling), RAG backends, document parsing pipelines, APIs and containerised services on Azure.
Build production AI applications and services: Design backend services, APIs, integrations, and LLM-powered agent workflows that solve problems for internal users.
Design, build, scale and maintain data pipelines that ingest, transform and manage data from embodied AI systems, experiments and operational environments.
Train, fine-tune, and evaluate computer vision models for tasks such as classification, object detection, segmentation, pose estimation, and action recognition.
Algorithmic Innovation: Spearhead research into optimal quantization methodologies for advanced low-precision data formats, driving the strategic adoption and expansion of our low-precision computing ecosystem to secure a competitive edge in numerical computing.
AI Search & AI Bot Team: Specializes in LLM application algorithm R&D for TikTok Search, covering core scenarios including AI Search Q&A cards and TikTok's official AI Assistant. We leverage LLM ChatBot, Agent, and RAG technologies to deliver accurate, real-time search responses for users, driving business growth through deep integration of technical innovation and business scenarios.
Lead and mentor a team of technical specialists while remaining hands-on with system architecture, establishing coding benchmarks, and guiding robust software design.