Ananya Shetty

Key Skills

Software

Labelbox

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code Programming

Text

Top Label Types

Computer Programming Coding

RLHF

Fine Tuning

Evaluation Rating

Function Calling

Prompt Response Writing SFT

Freelancer Overview

I am a software engineer with over three years of experience in backend development and hands-on work with AI and large language models (LLMs). My expertise includes fine-tuning LLMs, developing robust validation workflows, and designing prompt strategies to ensure highly accurate and reliable AI outputs. At Turing, I improved Java code generation and DSA-based model performance through targeted training, RLHF, and supervised learning, achieving up to 95% output accuracy. I am skilled in prompt engineering, LLM evaluation, and have strong experience with tools like ChatGPT, Claude AI, and data annotation techniques. My background in building secure, scalable systems and automating data workflows for leading financial clients has given me a solid foundation in handling sensitive data, ensuring compliance, and maintaining high data quality throughout the AI training lifecycle.

Labeling Experience

Software Engineer

LabelboxTextRLHFFine Tuning

• Developed validation strategies and evaluation workflows to achieve 95% output accuracy, ensuring production-level reliability. • Designed system prompts and agent behavior frameworks for LLM-based assistants, including tool-selection logic, function-calling strategies, error handling, and prevention of prompt injection in multi-turn AI conversations.

• Developed validation strategies and evaluation workflows to achieve 95% output accuracy, ensuring production-level reliability. • Designed system prompts and agent behavior frameworks for LLM-based assistants, including tool-selection logic, function-calling strategies, error handling, and prevention of prompt injection in multi-turn AI conversations.

2025

Software Engineer

LabelboxComputer Code ProgrammingComputer Programming Coding

• Fine-tuned LLMs to generate accurate Java solutions and improve DSA-based outputs by enhancing model training with backend and AI expertise. • Improved Java code generation precision by 30% and overall model performance by 25% using targeted fine tuning, RLHF, and supervised learning.

• Fine-tuned LLMs to generate accurate Java solutions and improve DSA-based outputs by enhancing model training with backend and AI expertise. • Improved Java code generation precision by 30% and overall model performance by 25% using targeted fine tuning, RLHF, and supervised learning.

2025 - 2025

Education

Work History