Harsh Wardhan
Identity: Academic AnalyticsOpen SourceStudent Utility

SRM Academic Suite

Deterministic CGPA/SGPA Grade Calculator & Target Planner

Engineering Challenge

“How do you build a grade calculator students actually trust?”

Products•Completed•Creator•2025
Student Users100+SRM University
Grade ParsingPDF.jsAutomated Extraction
Target EngineDeterministicZero Floating Errors
Client Side100%Privacy First
SRM Academic Suite
01 Product Context

Problem & Product Goals

University grading systems use weighted credit distributions that make manual GPA forecasting tedious and error-prone. The SRM Academic Suite provides an intuitive, deterministic web application that parses student grade cards and calculates target grade combinations needed to achieve specific cumulative GPAs.

Why I Built It

Many of my fellow students struggled to calculate how future semester grades would impact their cumulative GPA, often relying on imprecise estimates. I built a tool to give them exact, trustworthy mathematical calculations instantly.

02 Implementation

Technical Architecture & Core Engine

Core Engine Mechanism

Determines target letter grade matrices by performing constrained combinatorial search over available course credit weights.

Design Philosophy: Run 100% of calculations client-side to respect student data privacy while ensuring immediate computational feedback without server round-trips.

03 Technical Rigor

Engineering Decisions

Decision #01 • Deterministic Calculation vs Probabilistic Modeling
Problem:

Using heuristic AI models to forecast grades led to unexpected rounding discrepancies.

Decision:

Built a deterministic search algorithm mapping exact credit weightings directly to official university grading scales.

Tradeoff:

Requires maintaining updated credit-weight tables for different academic departments.

Outcome:

Achieved 100% mathematical accuracy across all CGPA/SGPA calculations.

04 Problem Solving

Challenges & Solutions

Parsing Unstructured PDF Grade Cards

Issue: Official university grade report PDFs vary significantly in layout and formatting between academic years.

Decision/Solution: Implemented robust layout-aware text token matching with fallbacks to extract course codes, credits, and letter grades reliably.

Future Roadmap

What's Next

Department-specific curriculum preset imports
Semester-over-semester performance visualizer
Mobile PWA offline installation support
Retrospective Summary

Key Takeaways

Takeaway #01

Deterministic mathematical tools build strong user trust when results match exact official calculations.

Takeaway #02

Client-side processing guarantees privacy and zero operational hosting costs for student utilities.

Continue Exploring

Related Engineering Projects