Ishant S.

Key Skills

Software

Other

Top Subject Matter

AI analytics chatbot and database-backed AI assistant

Productivity analytics data capture for AI analysis

Human activity recognition (deep learning on labeled dataset)

Top Data Types

Text

Top Task Types

Question Answering

Action Recognition

Freelancer Overview

Data Science Intern — Blackcoffer Pvt. Ltd. (AI chatbot development). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and Other. Education includes Bachelor of Technology, Madhav Institute of Technology and Science (2026). AI-training focus includes data types such as Text, Computer Code, and Programming and labeling workflows including Question Answering, Computer Programming, and Coding.

Labeling Experience

Data Science Intern — Blackcoffer Pvt. Ltd. (AI chatbot development)

TextQuestion Answering

Built a context-aware multilingual analytics chatbot with structured outputs and automated insights. Worked on generating forecasts and presenting information using charts and tables to support downstream interpretation. Contributed database-oriented tasks to support client issue resolution within an AI project environment. • Implemented backend FastAPI APIs for frontend integration. • Used SQL and database concepts for data retrieval and issue debugging. • Supported multilingual and voice-enabled user input handling. • Added comparative insight views with chart/table outputs.

2026 - Present

Software Developer Intern — Morgan Soft Innovations Pvt. Ltd. (monitoring system development)

Other

Developed a real-time employee monitoring system that captures productivity-related signals for analysis. Implemented secure REST APIs and interactive WebSocket communication to collect and transmit monitoring data. Although primarily an engineering role, the work supports AI-driven analysis by collecting screen, window, and activity logs as labeled-like inputs. • Captured screenshots, screen recordings, and window activity logs for productivity analysis. • Built a Python desktop application and integrated it with a Django web platform. • Implemented hierarchical user management with JWT authentication. • Engineered WebSocket-based handling of concurrent user sessions.

2025 - 2025

Project — Human Activity Recognition Using DeepConvLSTM

OtherTextAction Recognition

Trained deep learning models on the UCI HAR dataset for human activity classification. Performed data preprocessing and applied regularization (dropout, batch normalization) to improve generalization. Evaluated model quality using precision, recall, F1-score, and confusion matrices across activity classes. • Implemented CNN, LSTM, and hybrid DeepConvLSTM architectures. • Used stratified cross-validation for robust performance estimates. • Conducted model evaluation with confusion matrices and F1 metrics. • Worked with labeled activity categories from the dataset for training and testing.

2024 - 2024

Education

Work History