about skills projects competitive programming education contact resume ↗

Available for Opportunities

Pratham
Chelaramani

Final Year CS Student interested in Backend Engineering, Algorithms and System Design. I build things that scale.

Resume GitHub LinkedIn

pratham@portfolio ~ %

❯ cat profile.json

{

"name": "Pratham Chelaramani",

"role": "Backend Developer",

"location": "Gujarat, India",

"lc_rating": 2166,

"cf_rating": 1534,

"icpc_air": 256,

"stack": ["Python","Java","Go"],

"open_to_work": true

}

❯

01 // about

Who I am

I'm Pratham — a final-year Computer Science student at BVM Engineering College, Gujarat. My focus is on backend engineering: building systems that are fast, reliable, and battle-tested.

I enjoy going low-level. Whether it's architecting a Redis-inspired in-memory cache from scratch with 16-shard concurrency, or crawling the web at 2.26 pages/sec using Java's ExecutorService — I like knowing how things work under the hood.

Outside coursework, I compete seriously in competitive programming. I'm a Guardian on LeetCode and Specialist on Codeforces, and I've represented my college at ICPC Asia West Regionals.

Email Me LeetCode Codeforces

LeetCode

2166

Max Rating

🏅 Guardian

Codeforces

1534

Max Rating

🎖 Specialist

ICPC

AIR 256

Asia West Regionals

🌏 Amritapuri–Mysuru

Best LC Rank

GR 173

Biweekly 182

👥 ~28K participants

02 // skills

Tech Stack

Languages

Python Java Go SQL

Frameworks & Libraries

FastAPI Django NumPy Pandas

Databases

MySQL Redis MongoDB

Tools & Platforms

Docker Git GitHub AWS

03 // projects

Things I've Built

SpiderCore.java

SpiderCore

A high-performance multithreaded Java web crawler using ExecutorService & Phaser.

Achieved 2.26 pages/sec across 364 URLs via depth-limited BFS on Wikipedia
Thread-safe URL management with ConcurrentHashMap (O(1) dedup) & LinkedBlockingQueue
MySQL via JDBC for persistent storage with crash recovery across restarts
Real-time crawler metrics via AtomicInteger — crawled, failed, throughput

Java MySQL JDBC Concurrency

View Source

kv_cache.py

In-Memory Key-Value Cache

A Redis-inspired in-memory KV cache in Python with 16-shard architecture and a custom RWLock.

16-shard architecture with custom RWLock for parallel reads — zero external dependencies
DJB2 hashing + bitmask for O(1) shard selection; 700K entry cap with auto-eviction
Multi-stage Docker (non-root) with OS tuning for 1M+ file descriptor support
HTTP layer on FastAPI + Uvicorn with Pydantic validation & pytest coverage

Python FastAPI Docker Redis

View Source

04 // competitive programming