Ashpak Pendari

Neuralace AI Training Intern

GadagAI/Machine Learning
3roles
22skills
4education
2credentials

About

AI/Machine Learning Engineer with a strong passion for leveraging AI to solve complex real-world problems. Experienced in developing scalable solutions and optimizing models for production environments, with a focus on innovative AI advancements.

Experience

Neuralace AI Training Intern

AWIGN Enterprises Pvt. LTD.

Dec 2024 - Mar 2025

Collected and prepared diverse multimodal datasets (images, videos, and audio) to train AI systems for assisting visually impaired individuals, ensuring ethical compliance and high-quality annotations while applying data augmentation techniques to enhance model performance.

Freelancer

AI Agentic workflow for SOUL AI

Jan 2025 - Present

Working on Project Nebula, contributing to innovative AI solutions.

AI Agentic Workflow for SOUL AI

SOUL AI

Jan 2025 - Present

Working on Project Nebula.

Education

Karnataka State Rural Development And Panchayat Raj University Gadag

Master of Science, Computer Science (Data Analytics)

Dec 2022 - Nov 2024

JT College Gadag

Bachelor of Science

June 2017 - Sep 2021

Karnataka State Rural Development And Panchayat Raj University, Gadag

Master of Science, Computer Science (Data Analytics)

Dec 2022 - Nov 2024

JT College, Gadag

Bachelor of Science

June 2017 - Sep 2021

Skills

Data analysisData visualizationPythonPower BIMS ExcelAI prompt generationSQLPytorchAWSTensorFlowNeural NetworksAI tools specialistData Analysis and VisualizationWeka ExplorerMicrosoft ExcelCNNYOLOFlaskRandom ForestClusteringData PreprocessingMultimodal Dataset Preparation

Projects

IMDB Top 1000 Analysis

Conducted a comprehensive data analysis and visualization project on IMDB's Top 1000 movies, extracting insights into film trends and audience preferences.

Netflix Analysis

Performed exploratory data analysis and clustering of similar content by matching text-based features.

Rental Bike Regression Analysis

Engineered a customer churn prediction model using Random Forest, achieving a 90% accuracy rate and a 15% reduction in churn, validated through cross-validation and hyperparameter tuning. Deployed the model using Flask for real-time predictions.

Vehicle Taillight Detection Using Deep Learning

Developed a real-time deep learning model for vehicle taillight detection optimized for diverse conditions to enhance autonomous driving safety, improving accuracy from 85% to 92%.

Rural Immersion

Conducted rural surveys, data collection, analysis, and visualization, along with spreading awareness.