Projects & Capstones

Proof, not promises.

Every learner graduates with deployed systems and public write-ups. Here's what you'll ship.

The headliners

Three capstones, three levels

Track 1 Capstone

AI-Powered Multi-Tenant HRMS

Product Engineering

A production-deployed, multi-tenant HRMS with org hierarchy, RBAC and audit logs — plus an AI HR assistant, embeddings-based smart search and workflow suggestions. Real infra, real data flows, live demo.

Explore Track 1
Track 2 Capstone

Production-Grade Multi-Service Platform

Advanced Infrastructure

The HRMS extended into a multi-service platform: Kafka event streaming, SAML SSO, Elasticsearch search and WebRTC video — fully Kubernetes-deployed, observable and load-tested at 10k concurrent users.

Explore Track 2
Track 3 Capstone

Geo-Distributed Architecture Thesis

Internet Systems

A distributed system you design from scratch, deploy across 2+ regions, chaos-test and load-test at 100k+ users — then defend before an engineering panel and publish as a public architecture write-up.

Explore Track 3
Project showcase

Everything you'll build along the way

Track 1

Production CLI Tool

A real command-line tool shipped through a proper Git workflow — branches, PRs and peer review.

Track 1

Full Auth & Identity System

RBAC, email verification, JWT refresh, audit logging and Redis session caching.

Track 1

Admin Dashboard Frontend

Role-based views, data tables, approval flows and real-time notifications wired end-to-end.

Track 2

Distributed Notification System

Email + SMS + push via queues, with retry, dead-letter handling and delivery tracking.

Track 2

Realtime Activity Feed

Kafka event stream + Elasticsearch audit log + WebSocket live push to the dashboard.

Track 2

Production Infra on AWS

Terraform + Kubernetes with autoscaling, observability and blue-green CI/CD.

Track 3

HTTP/2 Server From Scratch

A minimal HTTP/2 server on raw TCP sockets — every frame, header and stream state.

Track 3

Raft Consensus Node

Leader election and log replication, verified on a 5-node cluster under failures.

Track 3

AI Infrastructure Layer

Self-hosted vLLM inference, HNSW vector search, a multi-agent workflow and a cost dashboard.

Build your portfolio piece

Start the track that matches your level and ship your first real system.

Explore Tracks

Ready to Build Something Remarkable?

From product discovery to deployment — our engineers and consultants are on hand to help you move faster.

No commitment required  ·  Response within 24 hours  ·  50+ projects delivered