For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
G
Gaurav M.

Gaurav M.

AI Engineer

India flagDelhi, India

Key Skills

Software

CVATCVAT

Top Subject Matter

Generative AI

Top Data Types

ImageImage
TextText
VideoVideo

Top Task Types

Question AnsweringQuestion Answering
Text GenerationText Generation
Object DetectionObject Detection
Text SummarizationText Summarization
MappingMapping
Fine-tuningFine-tuning
Bounding BoxBounding Box
PolygonPolygon
ClassificationClassification
SegmentationSegmentation
Point/Key PointPoint/Key Point

Freelancer Overview

During my tenure at Vdoit Technologies, my AI training and development experience has focused heavily on building, optimizing, and deploying generative AI and machine learning systems. As an AI/ML Intern, I designed and implemented a Retrieval-Augmented Generation (RAG) model, utilizing advanced embedding techniques to enable fast, accurate semantic search and context retrieval. I also developed a high-performance Q&A model using Sentence Transformers, optimizing retrieval and ranking mechanisms to achieve a real-time response latency of just 250 milliseconds. Transitioning into my role as an AI Engineer, I integrated OpenAI’s API into an EdTech platform's mathematics module to dynamically resolve student queries, successfully bridging advanced LLM capabilities with live applications.While my background is centered on engineering and architecture rather than manual data labeling, I have extensive experience managing the data pipeline foundational to AI training. I regularly utilize Python and AWS DynamoDB to extract, clean, and analyze large, complex student datasets to ensure accurate reporting and track engagement. This hands-on experience in processing data, fine-tuning retrieval mechanisms, and managing large-scale data manipulation allows me to effectively prepare data and optimize models for peak performance.

Labeling Experience

I annotated and categorized intent and prompt-response pairs to train and validate our custom Q/A models

I annotated and categorized intent and prompt-response pairs to train and validate our custom Q/A models. I also developed structured metadata tagging systems for text chunks to optimize our Retrieval-Augmented Generation (RAG) system, ensuring the retrieval mechanism matched queries with the correct context.I utilized these labeled datasets to train and optimize a transformer-based Q/A model using Sentence Transformers, fine-tuning the model's retrieval and ranking layers to bring response latency down to 250 milliseconds. Additionally, I handled data preprocessingextracting, cleaning, and filtering out noise from large educational datasets using Pythonto prepare high-quality pipelines for model ingestion and OpenAI API alignment.

Not specified

Education

D

Delhi Skills & Entrepreneurship UniversitY

Data Analytics

Data Analytics
Not specified

Work History

C

Company not specified

AI Engineer

Location not specified
Not specified