Pranav Krishna S N

•Project: Automated SDLC & Decision Pipeline using n8n

• Designed and implemented automated data-driven workflows using n8n to orchestrate SDLC stages and decision pipelines, improving operational efficiency and repeatability.

• Built agent-based automation systems to convert structured inputs (epics, JSON configurations) into deterministic backend architectures, enabling controlled experimentation and rapid prototyping.

• Developed automated data ingestion, parsing, validation, and normalization pipelines for AI agent outputs, ensuring high-quality data for downstream analytics.

• Integrated GitHub REST APIs as a data source and action layer to track repository states, commit SHAs, version history, and change logs for analytical workflow.

• Implemented error detection, recovery, and retry mechanisms to handle API conflicts, inconsistent states, and partial pipeline failures.

• Designed pipeline monitoring and control mechanisms including human-in-the-loop approvals, alerts, and checkpoints to ensure governance and reliability.

• Collaborated with cross-functional teams to align automation workflows with enterprise analytics, MLOps, and AI governance standards.

•Project: Datamat (Client Data Management Platform using Dremio Lakehouse)

•Completed a three-month internship contributing to the Datamat enterprise data platform, focused on client data management using Dremio lakehouse architecture.

•Designed and developed optimized virtual datasets and semantic views to support client-specific reporting, business intelligence, and analytics use cases.

•Gained hands-on experience in data virtualization, query optimization, schema design, and metadata management within enterprise-scale data environments.

•Engineered high-performance analytical data models to enable fast, reliable access across multiple heterogeneous data sources.

•Collaborated with cross-functional teams including data engineering, analytics, and business stakeholders to streamline data workflows and ensure consistency across client-facing dashboards.

•Worked within an enterprise lakehouse ecosystem integrating structured and semi-structured data for scalable analytics and reporting.

• Project: Computer Vision–Based Fruit Classification System
• Gained hands-on experience in Python programming with a focus on Artificial Intelligence and Machine Learning model development and experimentation.
• Built and trained deep learning models using Keras, leveraging NumPy for numerical computation and OpenCV for computer vision pipelines and image preprocessing.
• Designed and implemented an end-to-end image classification system to accurately classify multiple fruit categories from raw image datasets.
• •Engineered data preprocessing, augmentation, and feature extraction pipelines to improve model accuracy, robustness, and generalization.
• Evaluated and optimized model performance using training-validation pipelines, hyperparameter tuning, and performance metrics.
• Delivered a production-ready AI-driven computer vision solution aligned with real-world business use cases.