Waqar Ahmed

Key Skills

Software

Appen

Clickworker

CrowdSource

Data Annotation Tech

Google Cloud Vertex AI

Labelbox

Label Studio

Other

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Image

Text

Top Task Types

Bounding Box

Data Collection

Fine Tuning

Prompt Response Writing SFT

RLHF

Freelancer Overview

I am an experienced AI Data Trainer and Data Labeling Specialist with four years of expertise in training, evaluating, and optimizing AI models, including Large Language Models (LLMs). I have worked extensively on data annotation, model ranking, response evaluation, and AI safety testing, ensuring that machine learning models generate accurate, contextually relevant, and high-quality outputs. My role involved fine-tuning AI behavior, detecting hallucinations, transcribing multilingual content, and categorizing data for AI training, making me highly skilled in reinforcement learning (RLHF), supervised fine-tuning (SFT), and model validation. Beyond AI training, I bring a strong background in customer service operations and process optimization, having worked on large-scale projects with companies like Uber Eats, DoorDash, and Foodpanda. I have successfully led data-driven projects, quality audits, and AI-assisted content curation, ensuring seamless data enrichment and model improvement. My technical proficiency includes tools such as Salesforce, Zendesk, Jira, and other data management platforms, allowing me to bridge the gap between AI training and real-world applications. As a native Urdu and Punjabi speaker with advanced English proficiency, I am well-equipped for multilingual AI training and linguistic model enhancement.

Labeling Experience

Trajectory Based Work

OtherVideoText GenerationRLHF

I have actively contributed to trajectory-based AI training as an OpenAI Operator, focusing on refining LLMs (Large Language Models) through reinforcement learning and supervised fine-tuning (SFT). My work involved guiding AI behavior by generating high-quality training trajectories, ensuring that models learn patterns, context, and reasoning in a structured manner. Key tasks included: Creating and curating training dialogues to teach AI models natural, human-like interactions. Evaluating model responses and providing ranked feedback to improve coherence, factual accuracy, and ethical alignment. Identifying inconsistencies or biases in AI-generated text and refining datasets to enhance performance. Fine-tuning AI responses for improved decision-making by adjusting reward signals and reinforcement learning parameters.

I have actively contributed to trajectory-based AI training as an OpenAI Operator, focusing on refining LLMs (Large Language Models) through reinforcement learning and supervised fine-tuning (SFT). My work involved guiding AI behavior by generating high-quality training trajectories, ensuring that models learn patterns, context, and reasoning in a structured manner. Key tasks included: Creating and curating training dialogues to teach AI models natural, human-like interactions. Evaluating model responses and providing ranked feedback to improve coherence, factual accuracy, and ethical alignment. Identifying inconsistencies or biases in AI-generated text and refining datasets to enhance performance. Fine-tuning AI responses for improved decision-making by adjusting reward signals and reinforcement learning parameters.

2023 - 2024

Data Annotation & Model Ranking

OtherTextRLHFFine Tuning

Evaluating and ranking AI-generated responses for quality, factual accuracy, and contextual appropriateness.

Evaluating and ranking AI-generated responses for quality, factual accuracy, and contextual appropriateness.

2022 - 2023

Text Categorization & Sentiment Analysis

CrowdsourceTextBounding Box

Labeling and classifying textual data based on sentiment, intent, and relevance for AI model training.

Labeling and classifying textual data based on sentiment, intent, and relevance for AI model training.

2022 - 2022

Education

Work History