My contributions during my time at Draup live on my work GitHub: 👉 https://github.com/SamarthMahendra-Draup

My College: Github : 👉 [https://github.com/SamarthMahendraNEU ](https://github.com/SamarthMahendraneu

• 🦁 Northeastern University, Boston, MA
  Master of Science (MS), Computer Science
  Jan 2024 – Dec 2025
  Khoury College of Computer Sciences
  Relevant Courses:

  • CS 5010: Programming Design Paradigm
  • CS 5200: Database Management Systems
  • CS 5800: Algorithms
  • CS 6120: Natural Language Processing
  • CS 6140: Machine Learning
  • CS 5520: Mobile Application Development
  • CS 5500: Foundations of Software Engineering

• 🐍 Dayananda Sagar College of Engineering, Bangalore, India
  Bachelor of Engineering (BE), Computer Science
  Aug 2018 – Jul 2022 | CGPA: 8.59

• 👨‍💻 Associate Software Development Engineer – Backend
  Draup · Full-time
  Aug 2022 – Nov 2023 · 1 yr 4 mos · Bangalore (Hybrid)

  • Led platform modules for digital tech stack, outsourcing, and customer intelligence.
  • Revamped insights page for complex comparisons, boosting engagement by 40%.
  • Designed a dynamic query generation engine for chatbot pipelines — improved performance by 60% and cut dev time by 80%.
  • Migrated APIs to Elasticsearch for real-time aggregation (5× speedup).
  • Introduced advanced Boolean filter logic (e.g., (a AND b) OR c) for enhanced search.
  • Built subscription-based access control and enhanced platform performance (400% speedup, 50% cost reduction).
  • Reduced platform downtime from 4% to 1%; resolved issues with Datadog & AWS (75% faster).

• 🛠 Backend Engineering Intern
  Draup · Internship
  Apr 2022 – Jul 2022 · 4 mos · Bengaluru

  • Debugged API issues using Datadog.
  • Implemented caching for image requests to boost performance.
  • Created automated DB scripts to streamline data workflows.

• 🔬 Research Assistant (Patent Co-Inventor)
  Dayananda Sagar College of Engineering · Part-time
  Nov 2021 – Sep 2023 · 1 yr 11 mos · Bengaluru (Hybrid)

  • Project: Myocardium Wall Motion & Thickness Mapping (Patent Pending)
    App No: 202341086278 (India)
  • Developed novel image processing techniques for MRI cine scans.
  • Built algorithms for myocardium thickness + fibrosis mapping during cardiac cycles.
  • Optimized using NumPy, CuPy (GPU), multiprocessing with ThreadPoolExecutor.
  • Worked alongside clinicians from Sri Jayadeva Institute of Cardiovascular Sciences.

• 📊 JobStats.fyi — Real-Time Job Market Analytics
  Built a real-time analytics platform processing 15K+ submissions, reaching 1,000 sessions in the first 48 hours. Designed a distributed pipeline using Flask, MongoDB, Redis, Celery, and Prometheus/Grafana, with production monitoring and fault-tolerant background processing.

• 🧱 ButterDB — Custom Storage Engine (C)
  Implemented a key-value storage engine with on-disk B-Trees, fixed-size pages, write-ahead logging (WAL), crash recovery, and fine-grained locking. Focused on durability, concurrency control, and performance-critical storage internals.

• 🗄️ User-Space Unix File System (FUSE, C)
  Implemented a Unix-like file system in user space supporting core POSIX operations (read, write, mkdir, unlink, rename, chmod, stat). Designed on-disk structures including superblocks, bitmaps, inodes, and directory traversal over a disk image.

• 🧵 Concurrent Key–Value Store Server (C, pthreads, TCP)
  Built a multi-threaded TCP-based KV store with a listener/worker thread pool, thread-safe task queue, and monitor-based synchronization using mutexes and condition variables. Validated correctness with randomized concurrency tests.

• 🐚 Unix Shell Implementation (C)
  Implemented a Unix-style shell supporting pipes, I/O redirection, background execution, and conditional chaining (;, &&, ||). Built a custom parser and execution engine using fork(), exec*, waitpid(), and file descriptor manipulation.

• 🎧 Live Bluetooth Silent Disco
  Built a real-time system audio broadcast tool using WebSockets and Python, streaming low-latency audio to multiple Bluetooth clients via BlackHole.

• 🧠 LinkedInAssist — LLM-Powered Job Filter
  Chrome extension that filters LinkedIn job postings using natural-language prompts. Built with GPT-based inference and a Flask backend.

• 💬 Account Intelligence Chatbot (Hackathon @ Draup)
  Built an enterprise chatbot using RAG with LangChain, cross-encoder reranking, query expansion, Redis caching, and PostgreSQL-backed retrieval.

• 🧮 Java Portfolio Manager
  Designed an MVC-based investment simulator with external stock APIs, 100+ JUnit tests, and performance charts.

• 🚴‍♂️ Bike Rental Platform
  End-to-end BlueBikes-style system using React, Django REST, Redis, JWT authentication, and Azure deployment.

• 🩺 Myocardium Wall Motion Mapper (Patent Pending)
  Built a medical imaging pipeline to analyze cardiac wall motion from cine MRI scans using NumPy, multiprocessing, and interpolation.