For employers

Hire this AI Trainer

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

LoginCreate account
Invite to Job
Dinu T S

Dinu T S

Senior AI Trainer & Data Science Engineer (LLM, Computer Vision, SQL)

India flagThrissur, India
$90.00/hrExpertLabel StudioCVATLabelbox

Key Skills

Software

Label StudioLabel Studio
CVATCVAT
LabelboxLabelbox
ProdigyProdigy

Top Subject Matter

Healthcare & Life Sciences – Medical Imaging (DICOM) & Diagnostics
Technology & Software – Large Language Models (LLM) & Code Generation
Legal Tech & Compliance – Document Intelligence & NER

Top Data Types

Medical DicomMedical Dicom
VideoVideo
ImageImage

Top Task Types

Segmentation
Entity Ner Classification
RLHF

Freelancer Overview

Data Science Engineer in Contract Review, Compliance, and Legal Research. Brings 5+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Education includes Professional Certificate, Indian Institute of Management Kozhikode (2022) and Master of Science, Indian Institute of Technology Palakkad (2021). Well suited for text-focused AI training, including legal document review, compliance annotation, and rubric-based quality evaluation.

ExpertEnglish

Labeling Experience

LegalAssistant AI: Named Entity Recognition

DocumentEntity Ner Classification
Developed a specialized Named Entity Recognition (NER) framework for complex legal PDFs to power a RAG-based query system. Annotated thousands of legal clauses and entities to improve the model's contextual understanding of legal precedents. Engineered a "Context Injection" layer that utilized these labels to achieve an 85% improvement in answer accuracy compared to standard prompts. Provided a full technical "User Guide" as part of the project delivery for non-technical stakeholders.

Developed a specialized Named Entity Recognition (NER) framework for complex legal PDFs to power a RAG-based query system. Annotated thousands of legal clauses and entities to improve the model's contextual understanding of legal precedents. Engineered a "Context Injection" layer that utilized these labels to achieve an 85% improvement in answer accuracy compared to standard prompts. Provided a full technical "User Guide" as part of the project delivery for non-technical stakeholders.

2025 - 2025

Multi-Object Detection for Autonomous Systems

VideoBounding Box
Executed multi-object tracking and 2D/3D bounding box annotation for urban driving datasets to train obstacle detection models. Labeled and classified pedestrians, vehicles, and obstacles across diverse lighting and weather conditions. Implemented Python scripts to automate pre-labeling for high-confidence video frames, increasing throughput by 35%. Used SQL to query and segment specific image metadata to ensure a balanced and high-quality training set.

Executed multi-object tracking and 2D/3D bounding box annotation for urban driving datasets to train obstacle detection models. Labeled and classified pedestrians, vehicles, and obstacles across diverse lighting and weather conditions. Implemented Python scripts to automate pre-labeling for high-confidence video frames, increasing throughput by 35%. Used SQL to query and segment specific image metadata to ensure a balanced and high-quality training set.

2024 - 2024

LLM Reinforcement Learning (RLHF) for Code Generation

Computer Code ProgrammingRLHF
Participated in Reinforcement Learning from Human Feedback (RLHF) and Supervised Fine-Tuning (SFT) to optimize a large language model's coding capabilities. Evaluated and ranked model-generated Python and SQL snippets based on logical correctness, efficiency, and safety. Provided "Gold Standard" rewrite samples for complex technical queries, resulting in a 20% reduction in generated code bugs. Developed a custom tool to automate the generation of structured Markdown documentation from raw code.

Participated in Reinforcement Learning from Human Feedback (RLHF) and Supervised Fine-Tuning (SFT) to optimize a large language model's coding capabilities. Evaluated and ranked model-generated Python and SQL snippets based on logical correctness, efficiency, and safety. Provided "Gold Standard" rewrite samples for complex technical queries, resulting in a 20% reduction in generated code bugs. Developed a custom tool to automate the generation of structured Markdown documentation from raw code.

2022 - 2023

High-Precision Medical Image Annotation for Diagnostic AI

ImageClassification
Managed the annotation and classification of a high-resolution dataset of 25,000+ medical images (MRI and CT scans) to train a deep learning model for early-stage cancer detection.

Managed the annotation and classification of a high-resolution dataset of 25,000+ medical images (MRI and CT scans) to train a deep learning model for early-stage cancer detection.

2022 - 2023

Education

I

Indian Institute of Management Kozhikode

Professional Certificate, Data Science and Artificial Intelligence

Professional Certificate
2021 - 2022
I

Indian Institute of Technology Palakkad

Master of Science, Physics

Master of Science
2019 - 2021

Work History

B

Benzy Infotech

Data Science Engineer

Thrissur
2025 - Present
R

RP2

Data Engineer

Thrissur
2022 - 2025