For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
K
Ka K.

Ka K.

Xiaomi Automobile — Algorithm Research Intern (sensor failure anomaly grading via MLLM fine-tuning, data collection/feed

Hong Kong flagHongkong, Hong Kong

Key Skills

Software

Don't disclose

Top Subject Matter

Intelligent driving onboard sensor anomaly detection
LLM post-training for recommendation/pricing and negotiation social intelligence

Top Data Types

ImageImage
Computer Code ProgrammingComputer Code Programming
TextText

Top Task Types

Fine-tuningFine-tuning
RLHFRLHF

Freelancer Overview

Xiaomi Automobile — Algorithm Research Intern (sensor failure anomaly grading via MLLM fine-tuning, data collection/feed. Brings 1+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose. Education includes Master of Science, Tsinghua University (2021) and Bachelor of Science, Nanjing University (2021). AI-training focus includes data types such as Computer Code and Programming and labeling workflows including Fine-tuning and RLHF.

Labeling Experience

Taobao and Tmall Group (Xianyu) — Algorithm Research Intern (LLM/RL for pricing & social intelligence, benchmarks and offline evaluation)

Don't discloseRLHFRLHF

Interned at Taobao and Tmall Group (Xianyu) applying LLM post-training and reinforcement learning techniques to improve real-world pricing and social intelligence systems. For pricing, you explored LLM-based reasoning over product similarities and introduced reinforcement learning to incorporate pricing knowledge, then deployed the solution for A/B testing. For negotiation/social intelligence, you built an agent and developed an internal benchmark, then iterated strategy and dialogue using hierarchical reinforcement learning to improve offline performance. • Explored LLM-based intelligent pricing strategy with reinforcement learning • Designed an internal benchmark for agent iteration and evaluation • Applied hierarchical reinforcement learning to optimize strategy and dialogue • Deployed pricing solution for A/B testing and achieved revenue gains

2025 - 2025

Xiaomi Automobile — Algorithm Research Intern (sensor failure anomaly grading via MLLM fine-tuning, data collection/feedback loop)

Don't discloseFine-tuningFine-tuning

Interned as an Algorithm Research intern where you worked on an online sensor anomaly inspection system for intelligent driving using data collected and identified via online anomaly inspections. You optimized the model to correctly identify different anomaly types and produce accurate grading by fine-tuning a multimodal model. The work created a data feedback loop by collecting and filtering abnormal sensor data for subsequent iterations. • Fine-tuned a Qwen multimodal model for anomaly detection inspection • Designed/used data collection and filtering for abnormal sensor data • Conducted multi-task learning studies to improve detection capability • Performed offline evaluation to surpass the online inspection model

2025 - 2025

Education

N

Nanjing University

Bachelor of Science, Information and Communication Engineering

Bachelor of Science
2017 - 2021
T

Tsinghua University

Master of Science, Information and Communication Engineering

Master of Science
2021

Work History

X

Xiaomi Automobile

Algorithm Research Intern

Beijing
2025 - 2025
T

Taobao and Tmall Group

Algorithm Research Intern

Hangzhou
2025 - 2025