Engineering Distributed Systems & Data Pipelines.
I'm Harshita Loomba, a Computer Engineering student at Thapar Institute and Software Development Intern at PSPCL. I architect microservices platforms serving 10,000+ residents, build predictive ML pipelines, and optimize power grid distribution telemetry.
Capabilities & Impact
Engineering Highlights
Architecting microservices for 10,000+ residents, analyzing electrical telemetry across 12,477 feeders, and training predictive models.
Campus Residents Served
Architected an in-house microservices platform replacing paid third-party software, cutting institutional overhead by 40% and rendering 17 hostels in interactive Three.js 3D.
Power Feeders Analyzed
Engineered PHP/MySQL pipeline detecting 150+ false feeder clearances while computing optimal capacitor placement to cut transmission loss by 10%.
Customer Transaction Pipeline
Benchmarked LightGBM and Neural Networks on severe 10:1 imbalanced data using semi-supervised pseudo-labeling and RankGauss normalization.
AIESEC Team Leader
Outgoing Global Volunteer (oGV) Product Lead
Led a high-performing team of 15+ members while developing a data-driven dashboard analyzing 5 years of market trends, boosting approvals & revenue by 20% to earn national recognition.
Engineering Showcase
Featured Systems & Case Studies
Production microservices platforms, neural prediction pipelines, and automated cloud deployments.
HOSMAS — Hostel Management System
In-house microservices-based hostel management platform serving 10,000+ university residents across 5 operational modules, replacing a paid third-party institutional solution and reducing operational overhead by 40%.
- Architected an in-house microservices platform for 10,000+ residents across 5 operational modules, slashing third-party software overhead by 40%.
- Developed 6 independent Go and Django microservices with isolated PostgreSQL databases, consumed by 4 Next.js clients and secured via JWT authentication.
- Designed an interactive Three.js 3D visualization covering 17 hostels, enabling residents to navigate blocks, floors, and room layouts in real time.
Customer Transaction Prediction Engine
Data-driven predictive pipeline surfacing transaction behavior insights on an imbalanced dataset (~10:1 ratio), leveraging semi-supervised pseudo-labeling to boost AUROC from 0.89 to 0.95.
- Built a data-driven pipeline on a severely imbalanced dataset (~10:1 class ratio) using semi-supervised pseudo-labeling to leverage unlabeled records, improving model AUROC from 0.89 to 0.95.
- Benchmarked LightGBM, neural networks, and ensemble models via 10-fold cross-validation with RankGauss normalization and shuffle augmentation.
- Extracted high-fidelity feature importances to identify key driver metrics governing transactional velocity.
LeetCode Progress Tracker & CI/CD
Containerized problem-solving tracker with topic-wise filtering, progress analytics, and automated Jenkins CI/CD pipeline serving 100+ active users.
- Built and deployed full-stack coding tracker using React.js, Express.js, and MongoDB with RESTful APIs, enabling topic-wise filtering and analytics for 100+ active coders.
- Containerized application with Docker and established automated CI/CD via Jenkins, cutting release times by 30% and eliminating manual steps.
Banking Client Analytics Dashboard
Interactive Power BI and Python analytics dashboard tracking client lending, deposit velocity, and engagement across multi-tiered customer relationships.
- Constructed interactive Power BI dashboard tracking loan, deposit, and engagement metrics cross-filtered by relationship, nationality, and income tier.
- Implemented calculated fields (Engagement Timeframe, Income Band, Total Loan, Total Deposit) converting raw transaction logs into decision-ready metrics.
Technical Arsenal
Skills & Core Competencies
Full-stack microservices, cloud infrastructure, machine learning toolkits, and agile team execution.
Programming Languages
Systems, scripting, and web languages
Frameworks & Libraries
Frontend, backend microservices, and ML models
Tools, Cloud & Databases
Databases, containerization, CI/CD, and infrastructure
Core Competencies
Leadership & delivery
Career & Impact
Work Experience & Leadership
Enterprise grid pipeline optimization at PSPCL and youth talent leadership at AIESEC.
PSPCL
Software Development Intern
- Engineered Fake Closure Detection Pipeline: Built a PHP and MySQL-based system to flag fraudulent feeder closure events, detecting 150+ false clearances and reducing report backlog by 15%.
- Optimized Grid Capacitor Placement: Analyzed half-hourly power consumption data across 12,477 distribution feeders, identifying optimal capacitor placement projected to reduce transmission losses by 10%.
AIESEC in Patiala, AIESEC in India
Team Leader, Outgoing Global Volunteer
- Led a high-performing team of 15+ members managing the Outgoing Global Volunteering (oGV) product for 120+ individuals, driving strategic alignment and performance management to achieve organizational objectives.
- Developed a data-driven dashboard analyzing 5 years of market trends to identify high-potential markets, and executed targeted marketing strategies that increased approvals and revenue by 20%, earning national recognition.
Thapar Institute of Engineering and Technology
Bachelor of Engineering in Computer Engineering • Patiala, Punjab
Credentials & Recognition
Certifications & Achievements
Industry simulation credentials, cloud certifications, and technical specializations.
AWS Educate — Generative AI
Certified in AWS Educate: Introduction to Generative AI, demonstrating foundational understanding of LLMs, foundation models, prompt engineering, and cloud AI use cases.
Deloitte Australia — Data Analytics
Completed professional job simulation on Forage, applying data analysis, forensic dashboarding, and business intelligence to solve enterprise operational challenges.
Thapar Summer School 2024
Completed intensive training in Python-based Machine Learning and Deep Learning, covering CNNs, NLP transformers, and advanced image processing.