Sabin Pant
An hour of planning saves a week of debugging. Before I even fire up my IDE, I'm mapping out system flows, sketching schemas, and predicting where things might break (hello, race conditions). For me, the design phase isn’t just prep work—it’s the actual work.
The developer behind the architecture.
Backend-first. System-design obsessed. Built to handle what happens at scale.
Available for opportunities
Full-time · Internship · Freelance
Origin Story
It started with a game. I wanted to build one, so in high school, I decided to pursue a Computer Science path. I began by picking up C and C++, and that first taste of building a powerful foundation in programming paradigms, Data Structures, and Algorithms (DSA) was addictive. As my curiosity deepened, my interests naturally shifted from games to real-world applications. Every layer I pulled back revealed something more interesting underneath, leading me from foundational systems into backend architecture, robust data pipelines, and scalable cloud infrastructure.
Why Backend & System Design
Backend is where the abstract logic maps neatly to true systemic resilience. I'm drawn to it because that's where the architectural complexity lives, handling race conditions, ensuring transaction atomicity, designing robust relational schemas, and preserving consistency across distributed layouts. I live for mapping actor interactions, visualizing cross-functional data flows, and stress-testing edge cases on paper long before opening an IDE. The frontend captures attention, but the backend earns trust.
Where I'm Headed
My objective is to build enterprise-grade critical infrastructure, the type that demands absolute zero-fault execution: banking networks, high-reliability payment systems, real-time transaction clearing environments, and low-latency data pipelines. I operate best in environments where a bug isn't just an interface flaw, but an engineering liability. That high-stakes paradigm motivates me to design for predictable consistency, reliable correctness, and structural scale.
Gaming
Gaming is what got me into coding in the first place. I started out wanting to build my own games, and that curiosity led me down the rabbit hole of programming. These days I still enjoy diving into a good game like Mobile Legends or sometimes PUBG, though I'm not as active as I used to be.
Fun Fact
I once spent more time modding a game than actually playing it
What I work with.
Backend
6 skills
Frontend
5 skills
Databases
6 skills
Cloud & Infra
6 skills
Architecture
6 skills
Tools & Workflow
5 skills
* AWS Certified: Cloud Practitioner, Solutions Architect, Developer, SysOps, ML Specialty
Where I have worked.
Full-Stack Developer Intern
Leaflet Digital Solutions
Kathmandu, Nepal
Architecture & implementation ownership
Building data-intensive systems with Node.js, Express, Next.js and PostgreSQL — responsible for both architecture decisions and day-to-day implementation.
Secured API pipelines at scale
Designed and hardened API layers with JWT authentication and role-based access control across multi-tenant applications.
Zero-ambiguity handoffs
Produced full Swagger API documentation, cutting frontend integration time and eliminating back-and-forth over contract details.
Next chapter
Open to full-time roles, internships & freelance
Things I have built.
Every project starts with architecture. Here is what that looks like in practice.
Hospital Management System
MediLife
A full-featured hospital management platform built under strict MVC architecture covering patient registration, doctor approval workflows, appointment scheduling, billing, and medical records across three roles.
- Dual booking flow: Quick Consult and Admin-assigned appointments with rejection handling
- BCrypt password hashing, session-based auth with configurable timeout and account lock/unlock
- Notification system triggered by booking, billing, and account events with CSV export
Production-Grade Real-Time Chat Application
Nebula Chat
A full-stack, production-deployed chat platform with passwordless OTP login, Google OAuth 2.0, one-to-one and group messaging, typing indicators, online presence, and unread badges — built end-to-end under strict clean architecture with zero layer bleeding, deployed across Vercel, Render, Neon, and Upstash.
- Dual auth flows: passwordless OTP email login and Google OAuth 2.0 with server-side audience validation — JWT access tokens (15 min, in-memory) paired with httpOnly 7-day refresh cookies and full token rotation on every refresh
- Socket.IO real-time engine with typing indicators, online presence, unread badge counts, and read receipts — all synced across multi-device sessions via a JWT-authenticated singleton SocketManager
- 3NF-normalized PostgreSQL schema across 8 tables with composite PKs on junction tables, soft-delete participant tracking (leftAt), and Redis pub/sub for scalable event broadcasting
- Three-layer clean architecture enforced end-to-end: controllers route only, services own all business logic, repositories handle all data access — with Zod schemas validating every API boundary and typed HttpException subclasses caught by a single global error middleware
- Deployed stack: React + Vite on Vercel, Express + Socket.IO on Render, Neon serverless PostgreSQL, Upstash serverless Redis, and Resend for transactional email — with Docker Compose for local parity
Network Intrusion Detection & Smart Data Analysis
NetGuard IDS
An end-to-end data science pipeline applied to a real-world IDS network traffic dataset of 488K records and 80 features. Covers full data preparation, statistical analysis, correlation discovery, and hypothesis-driven classification of BENIGN vs. Infiltration traffic.
- Full data wrangling pipeline: NaN removal, duplicate detection (28.56%), feature selection down to 16 key columns
- Pearson correlation heatmap revealing 0.9951 overlap between Packet Length Mean & Average Packet Size — flagging redundant features before modelling
- Welch's t-test on Flow Duration (p=0.4736) proving single-feature classification is insufficient, motivating multi-feature ML approach
Credentials I have earned.
Industry-recognised certifications across cloud architecture, machine learning, and product design — each linked to its verified credential.
AWS Cloud Foundations
Core cloud concepts, AWS global infrastructure & core services.
AWS Data Engineering
Data lakes, ETL pipelines, and analytics services on AWS.
AWS Academy
AWS ML Foundations
Machine learning concepts, SageMaker, and model lifecycle.
AWS Academy
AWS ML for NLP
Natural language processing with Comprehend & Lex.
AWS Academy
AWS Generative AI
Foundations of GenAI, Bedrock, and prompt engineering.
AWS Academy
Let us connect.
Whether you have a project in mind, a job opportunity, or just want to say hi my inbox is always open.
Get in touch
I am currently open to internship and full-time opportunities. If you have a role that matches my profile or want to collaborate on something exciting, reach out!