Garv A.

Key Skills

Software

Internal/Proprietary Tooling

Top Subject Matter

Hotel OTA data collection for analytics/ML

LLM RLHF training data curation and evaluation

Top Data Types

Text

Image

Top Task Types

Data Collection

Fine-tuning

RLHF

Function Calling

Prompt + Response Writing (SFT)

Freelancer Overview

AI Developer, Blue-Enigma (Early-Stage SaaS Startup). Brings 3+ years of professional experience across complex professional workflows, research, and quality-focused execution. Education includes Bachelor of Technology, Ajeenkya D Y Patil University (2024) and Higher Secondary School Diploma, Noble International School (2020). AI-training focus includes data types such as Text and Image and labeling workflows including Data Collection and Fine-tuning.

Labeling Experience

AI Developer, Blue-Enigma (Early-Stage SaaS Startup)

Don't discloseTextComputer Programming/Coding

Built and operated an OTA data ingestion pipeline by scraping multiple hotel sources to collect structured inventory and pricing information for downstream analytics. Normalized and organized hotel-level datasets including availability trends, amenities, and cancellation policies to support reliable model-ready inputs. Implemented validation automation via an AI-assisted extraction agent to ensure schema consistency and reduce anomalies during high-volume ingestion. • Scraping-driven dataset collection across Agoda and Booking.com • Data normalization for structured hotel records • Automated response validation and anomaly detection • Resilient ingestion with adaptive retry/throttling/backoff

2025 - 2025

LLM-Python Engineer, Turing

Don't discloseTextRLHF

Improved LLM performance using reinforcement learning from human feedback (RLHF) to enhance contextual accuracy and response alignment. Curated and cleaned 50K+ text samples across multiple domains to strengthen robustness and training quality. Partnered with data science and research teams to define evaluation metrics and automate model-comparison dashboards for iterative tuning. Integrated transformer-based models into production inference workflows and deployed them via Flask microservices. • RLHF-based model improvement • Text dataset curation and cleaning (50K+ samples) • Automated evaluation and model comparison for tuning • Production deployment of inference endpoints

2024 - 2024

Education

Work History