Computer Science undergraduate at IIIT Kota, interested in software engineering, systems programming, AI/ML, and competitive programming.
I enjoy building systems from the ground up, working close to performance-critical code, and applying machine learning to practical problems.
- 🏆 ICPC Asia Regionalist 2024
- 💻 900+ algorithmic problems solved
- ⭐ 4★ CodeChef | Codeforces Specialist
- 🤖 Selected for Amazon ML Summer School 2025 & 2026
- 🔧 Currently exploring C++ systems programming, debugging, compilers, and low-level software
- Building systems projects in modern C++
- Learning more about compilers, debuggers, Linux internals, and computer architecture
- Developing backend and AI-powered applications
- Practicing advanced Data Structures & Algorithms
- Contributing to open-source projects
- Modern C++17/20 and systems programming
- Linux process and memory internals
- Compilers, LLVM, debugging, and executable formats
- Backend and distributed systems
- Machine Learning, RAG, and LLM-based applications
Languages: C++, Python, Java, SQL, Bash
Systems & Core CS:
Data Structures & Algorithms, Operating Systems, Computer Networks, DBMS, OOP, Multithreading, Socket Programming, Linux
Backend:
FastAPI, REST APIs, PostgreSQL, SQLite, Redis, AsyncIO
AI / ML:
Scikit-Learn, XGBoost, TensorFlow, PyTorch, OpenCV, RAG, LangChain, Sentence Transformers, FAISS, pgvector
Tools:
Git, GitHub, Docker, CMake, CI/CD, Linux, Postman
Cloud:
Google Cloud, Microsoft Azure
C++17 • Linux • ptrace • ELF • x86-64
A lightweight Linux debugger built from scratch using ptrace, supporting software breakpoints, single-step execution, register inspection, memory access, and ELF symbol lookup.
Demonstrates low-level process control, CPU registers, executable formats, and debugging internals.
C++17 • Multithreading • POSIX Sockets • TCP/IP • CMake
A Redis-inspired storage engine with a multi-threaded server, O(1) key lookup, LRU eviction, TTL expiration, AOF persistence, and crash recovery.
- 50K+ operations/sec
- Thread-safe concurrent access
- RESP-style protocol parsing
Python • FastAPI • PostgreSQL • pgvector • LangChain
An LLM-powered document intelligence system combining dense vector retrieval with PostgreSQL full-text search and cross-encoder reranking.
- Handles 100+ page documents
- Hybrid semantic + keyword retrieval
- LLM evaluation for faithfulness and relevance
Python • XGBoost • Scikit-Learn • Optuna • Pandas
Built an end-to-end ML pipeline on 50K+ records and 105 features, achieving 9.52% MAPE using feature engineering, cross-validation, and hyperparameter optimization.
- ICPC Asia Regionalist 2024 — Ranked 205th onsite among 500+ teams
- Amazon ML Summer School 2025 & 2026 — Selected among the top applicants nationwide
- CodeChef 4★ | Codeforces Specialist
- Solved 900+ algorithmic problems
- Top 0.5% — IIT Kharagpur National Data Science Hackathon
- Top ~3% — Adobe GenSolve 2025
- Flipkart GRiD 7.0 National Semi-Finalist
- Hacktoberfest 2023 — Contributed across multiple open-source repositories