Available for Opportunities

M Geshta
Berling

M.Tech CSE · Amrita Vishwa Vidyapeetham

GATE-qualified M.Tech student and co-developer of FuelFlow — a live enterprise SaaS CRM & ERP platform built with React, TypeScript, and Node.js. Published researcher at ACM and IEEE international conferences. Experienced in building full-stack web applications, Python desktop tools, and computer vision systems.

8.88 B.Tech CGPA
2× Published Papers
Live SaaS Product
GATE Qualified

Where I've Built

2024 — Present
Freelance / Co-Developer

FuelFlow CRM & ERP Platform

Independent Software Developer
React TypeScript Node.js Express.js PostgreSQL GitHub Actions CI/CD RBAC
  • Co-designed and developed a full SaaS CRM and ERP platform serving real clients, covering sales pipeline, HR attendance, salary management, inventory tracking, dispatch, AMC, and after-sales service ticketing — live in production.
  • Architected the PostgreSQL schema and implemented JWT-based authentication with role-based access control (RBAC), isolating admin, manager, and service technician privilege levels.
  • Built responsive React/TypeScript UI modules backed by RESTful Express.js APIs; configured GitHub Actions CI/CD pipelines for automated linting, testing, and cloud deployments.
2025
Freelance Client Project

Milk Delivery Management App

Freelance Software Developer
Python CustomTkinter pandas openpyxl PyInstaller Inno Setup
  • Delivered a production-ready Windows desktop application for a dairy business client, managing daily delivery records, billing calculations, and auto-generated customer invoices.
  • Built spreadsheet automation with pandas and openpyxl to compile formatted monthly/yearly financial reports; integrated automated WhatsApp billing notification templates.
  • Packaged and deployed the application as a secure single-file Windows installer using PyInstaller and Inno Setup with data persistence across updates.

What I've Built

ACM Published

Criminal Finder — Dual-Mode AI Detection System

A dual-mode criminal identification platform with a MERN stack web interface for records management and a real-time live video monitoring mode using OpenCV for frame-by-frame face detection with an automated alert mechanism on positive matches.

Research presented at STAI 2026 · ACM Fremont Chapter, USA (2nd Int'l Conference on Smart Technology & AI)
MERN Stack OpenCV Computer Vision Real-Time Detection REST API
View on GitHub
IEEE Published

Power Bank from E-Waste & Used Batteries

Designed and built a cost-effective, eco-friendly power bank by repurposing discarded Lithium-ion battery units, addressing critical e-waste reduction. Engineered custom safety-compliant charging/discharging circuits with over-discharge protection and dual-output support.

Research presented at ICSCSS 2025 · IEEE Systems Council, Coimbatore
Circuit Design Lithium-Ion E-Waste Reuse Hardware
View Certificate

My Toolkit

Languages
Java Python JavaScript TypeScript HTML5 CSS3
Frontend
React TypeScript Bootstrap Responsive Design CustomTkinter
Backend
Node.js Express.js REST APIs JWT Auth RBAC
Database & Data
PostgreSQL MariaDB pandas openpyxl JSON Excel Automation
Tools & DevOps
Git GitHub GitHub Actions (CI/CD) VS Code Antigravity PyInstaller Inno Setup
AI / Vision / Concepts
OpenCV Machine Learning Agile / SDLC System Design DSA OOP
AI Dev Tools:
GitHub Copilot Claude (Antigravity) Gemini Google AI Studio Cursor AI

Academic Background

Present

M.Tech in Computer Science & Engineering

Amrita Vishwa Vidyapeetham, Coimbatore

Admitted via GATE Qualification
2022 — 2026

B.Tech in CSE (AIML)

Kalasalingam Academy of Research & Education, Virudhunagar

CGPA: 8.88
STAI 2026 — ACM Fremont Chapter, USA
Research Publication · Criminal Finder System
ICSCSS 2025 — IEEE Systems Council
Research Publication · E-Waste Power Bank
C Programming & Data Structures
CodeChef Certification
Database Management Systems
CodeChef Certification

Let's Build
Something
Together.

Open to full-time roles, internships, freelance contracts, and research collaborations. I typically respond within 24 hours.