Jl Z.

Key Skills

Software

Other

Top Subject Matter

Customer Support Automation

Top Data Types

Text

Image

Top Task Types

Fine-tuning

Freelancer Overview

LLM Fine-tuning and Evaluation for Intelligent Chatbot. Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Master of Science, Stanford University (2024) and Bachelor of Science, University of California, Berkeley (2024). AI-training focus includes data types such as Text and labeling workflows including Fine-tuning.

Labeling Experience

LLM Fine-tuning and Evaluation for Intelligent Chatbot

OtherTextFine-tuning

Fine-tuned the Llama-3 8B model using LoRA to improve performance for customer support tasks. Evaluated the model output via RAG and incorporated human preference studies into the training process. Focused on optimizing model answers and reducing hallucination rates in real-world settings. • Applied efficient fine-tuning strategies to adapt the LLM to a domain-specific context. • Used retrieval-augmented generation (RAG) for better answer accuracy and grounding. • Included human evaluators for systematic feedback and preference scoring. • Addressed common LLM pitfalls such as irrelevant or hallucinated information.

2024 - 2024

Education

Work History