Nandakishor T V

Key Skills

Software

CVAT

Labelbox

Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Top Label Types

Emotion Recognition

Evaluation Rating

Question Answering

Red Teaming

RLHF

Translation Localization

Freelancer Overview

I am an AI data evaluator with a strong financial background and over five years of experience in retail banking operations, specializing in financial compliance, credit risk assessment, and fraud detection. My expertise lies in providing high-quality ground truth data and meticulously labeling and annotating financial datasets to train and evaluate AI models in the banking and credit domains. I have hands-on experience with LLM evaluation, particularly in financial localization and multilingual alignment (English-Malayalam), ensuring that complex regulatory and compliance terminology is accurately represented in AI outputs. My work includes rubric-based scoring, auditing model reasoning for explainability, and adversarial testing to identify and prevent model hallucinations. I am passionate about building reliable, unbiased AI systems that meet strict compliance standards and deliver trustworthy results in real-world financial applications.

Labeling Experience

AI Data Specialist (Localization & RLHF Evaluator)

CVATTextRLHF

As an AI Data Specialist (Localization & RLHF Evaluator), I evaluated conversational outputs for Large Language Models (LLMs) with a focus on English and Malayalam. My tasks included localization, prompt rewriting for empathy, and auditing model reasoning for cultural accuracy. I ensured outputs complied with empathy and Trust & Safety guardrails.• Evaluated LLM responses for cultural and emotional appropriateness via RLHF• Rewrote robotic prompts to make outputs empathetic and human-centric• Audited model outputs for intent and accuracy, guarding against hallucinations• Scoped work for English-to-Malayalam language localization in AI systems

As an AI Data Specialist (Localization & RLHF Evaluator), I evaluated conversational outputs for Large Language Models (LLMs) with a focus on English and Malayalam. My tasks included localization, prompt rewriting for empathy, and auditing model reasoning for cultural accuracy. I ensured outputs complied with empathy and Trust & Safety guardrails.• Evaluated LLM responses for cultural and emotional appropriateness via RLHF• Rewrote robotic prompts to make outputs empathetic and human-centric• Audited model outputs for intent and accuracy, guarding against hallucinations• Scoped work for English-to-Malayalam language localization in AI systems

2025

Education

Work History