Jay Nagose

Key Skills

Software

Other

OpenCV AI Kit (OAK)

Top Subject Matter

Indian Income Tax law

Legal/NLP LLMs

Material Volume Measurement

Top Data Types

Text

Image

Document

Top Task Types

Fine-tuning

Segmentation

Entity (NER) Classification

Freelancer Overview

LLM Fine-Tuning/Data Annotation for Legal Compliance AI. Brings 3+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Other and OpenCV AI Kit (OAK). Education includes Bachelor of Technology, Sanjivani College of Engineering, Kopargaon (2022) and Higher Secondary Certificate, Saibaba K.V.M. and Jr.College, Shirdi (2022). AI-training focus includes data types such as Text and Image and labeling workflows including Fine-tuning, Segmentation, and Entity (NER) Classification.

Labeling Experience

LLM Fine-Tuning/Data Annotation for Legal Compliance AI

OtherTextFine Tuning

I fine-tuned the LLaMA 3.2-8B large language model on a curated dataset of Indian tax legal documents to adapt it for Indian Income Tax compliance queries. The process involved preparing, curating, and annotating legal texts and statutory documents to improve LLM legal advisory accuracy. My work centered on domain adaptation of instruction-following models for retrieval-augmented generation and factual precision. • Curated and annotated domain-specific legal texts for LLM training • Implemented QLoRA-based fine-tuning pipelines for language model adaptation • Evaluated and measured improvements in citation accuracy and context retrieval • Integrated with RAG pipelines and FAISS for relevant, context-aware query results

I fine-tuned the LLaMA 3.2-8B large language model on a curated dataset of Indian tax legal documents to adapt it for Indian Income Tax compliance queries. The process involved preparing, curating, and annotating legal texts and statutory documents to improve LLM legal advisory accuracy. My work centered on domain adaptation of instruction-following models for retrieval-augmented generation and factual precision. • Curated and annotated domain-specific legal texts for LLM training • Implemented QLoRA-based fine-tuning pipelines for language model adaptation • Evaluated and measured improvements in citation accuracy and context retrieval • Integrated with RAG pipelines and FAISS for relevant, context-aware query results

2025 - Present

Computer Vision Segmentation/Annotation for Automated Volume Measurement

Opencv AI Kit OakImageSegmentation

I fine-tuned instance segmentation models (SAM-2) and vision LLMs to detect and measure material volumes using computer vision for industrial use cases. My role encompassed preparing annotated image data, implementing 3D reconstruction pipelines, and validating measurement accuracy. The annotation work aimed to enable precise automated detection for material volume estimation in an industrial setting. • Prepared, annotated, and labeled image datasets for segmentation tasks • Applied segmentation models and evaluated their performance on new image data • Enabled accurate and automated measurement of material via dual-camera vision • Reduced manual measurement time and improved error margins for industrial clients

I fine-tuned instance segmentation models (SAM-2) and vision LLMs to detect and measure material volumes using computer vision for industrial use cases. My role encompassed preparing annotated image data, implementing 3D reconstruction pipelines, and validating measurement accuracy. The annotation work aimed to enable precise automated detection for material volume estimation in an industrial setting. • Prepared, annotated, and labeled image datasets for segmentation tasks • Applied segmentation models and evaluated their performance on new image data • Enabled accurate and automated measurement of material via dual-camera vision • Reduced manual measurement time and improved error margins for industrial clients

2024 - Present

RL-Based Anomaly Detection Data Annotation (Copyright Registered)

OtherTextFine Tuning

I developed reinforcement learning-based systems for anomaly detection in network traffic by labeling and classifying diverse network behavior datasets. The work involved designing, annotating, and preparing training data for RL-based anomaly detectors and evaluating their performance. Advanced RL approaches (DQN, PPO, HER) were implemented to improve detection precision over traditional systems. • Labeled network traffic data for use in RL anomaly detection • Created and maintained diverse data samples for training neural networks • Evaluated detection systems for accuracy and performance against baselines • Documented and registered methods through copyright for reproducibility

I developed reinforcement learning-based systems for anomaly detection in network traffic by labeling and classifying diverse network behavior datasets. The work involved designing, annotating, and preparing training data for RL-based anomaly detectors and evaluating their performance. Advanced RL approaches (DQN, PPO, HER) were implemented to improve detection precision over traditional systems. • Labeled network traffic data for use in RL anomaly detection • Created and maintained diverse data samples for training neural networks • Evaluated detection systems for accuracy and performance against baselines • Documented and registered methods through copyright for reproducibility

2025 - 2025

AI-Powered Food and Nutrition Label Annotation (DIPEX Project)

OtherTextEntity Ner Classification

I developed and trained an AI-powered nutrition app using OCR and large language models for instant analysis of food labels and nutrition tables. The project required collecting, annotating, and labeling food label text data for NER and classification tasks. This work enabled automated extraction of ingredient and nutrition details for use in dietary safety alerts and personalized recommendations. • Collected and annotated food label text and nutrition data • Implemented OCR pipelines to automate data extraction • Used LLMs for text analysis and entity recognition • Provided accurate ingredient classification and safety alerts for end users

I developed and trained an AI-powered nutrition app using OCR and large language models for instant analysis of food labels and nutrition tables. The project required collecting, annotating, and labeling food label text data for NER and classification tasks. This work enabled automated extraction of ingredient and nutrition details for use in dietary safety alerts and personalized recommendations. • Collected and annotated food label text and nutrition data • Implemented OCR pipelines to automate data extraction • Used LLMs for text analysis and entity recognition • Provided accurate ingredient classification and safety alerts for end users

2025 - 2025

Education

Work History