Amar P.

Key Skills

Software

AWS SageMaker

Anno-Mage

Appen

Argilla

Axiom AI

CVAT

Data Annotation Tech

Labelbox

Label Studio

Roboflow

Clickworker

CloudFactory

CrowdSource

Dataloop

Datumbox

Datasaur

Datature

Dataturk

Deep Systems

Diffgram

Doccano

Encord

Figure Eight

Humanatic

Other

Top Subject Matter

LLM evaluation for factual accuracy

Safety Domain Expertise

policy compliance

Top Data Types

Text

Image

Audio

Top Task Types

RLHF

Freelancer Overview

Analyst - AI/LLM Practice. Brings 4+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include N and A. Education includes Bachelor of Technology, Sant Gadge Baba Amravati University (2023). AI-training focus includes data types such as Text and labeling workflows including Evaluation, Rating, and RLHF.

Labeling Experience

Analyst - AI/LLM Practice

Text

Evaluated, ranked, and refined AI-generated responses to complex prompts while enforcing quality and compliance requirements. Conducted safety and policy compliance reviews to filter harmful content, bias, and hallucinations. Provided structured feedback and rationales to support iterative model training and fine-tuning. • Performed factuality and instruction-following checks on high-volume outputs • Analyzed structural anomalies, logical fallacies, and grammatical/syntax issues • Assessed tone appropriateness and adherence to strict project guidelines • Measured productivity against weekly AHT and quality targets by meeting/exceeding KPIs.

2026 - 2026

Associate Process Executive - AI Data Annotation & LLM Evaluation (Client: Nvidia Graphics Pvt. Ltd.)

TextRLHF

Created, labeled, and validated high-quality multimodal training datasets to support next-generation generative AI model development. Applied taxonomy alignment rules and annotation guidelines to audit datasets, resolve edge cases, and reduce labeling errors. Audited model outputs for human-value alignment to identify subtle hallucinations and policy violations, supporting model safety. • Labeled and validated multimodal data (text, image, video, audio) per annotation guidelines • Performed prompt engineering and iterative response refinement to improve output quality scores • Conducted quality audits achieving 98% while meeting SLA benchmarks • Performed edge-case identification and quality-driven dataset improvements for training

2024 - 2026

Education

Work History