Samarth Mahendra
"Turning code into poetry. Automating the tedious. Designing for the next billion users."
About Me
A glimpse into my professional passions and personality.
Software Engineer passionate about backend systems, distributed architecture, and AI integrations.
Currently working on
A distributed job tracking system (JobStats) scraping 50,000+ records daily using Celery, Redis, Prometheus, and Puppeteer.
Looking to collaborate on
LLM-powered productivity tools, scalable backend infrastructure, or real-time data processing systems.
Ask me about
How I improved chatbot performance by 60%, or optimized API costs using query partitioning and dynamic query generation.
Fun fact
I skated 22.3 km in a single session! 🛹
Education
Academic background and coursework.
Northeastern University
Master of Science in Computer Science · Jan 2024 – Dec 2025
- Concentration: Artificial Intelligence, Data Science, Systems & Software
- Relevant Coursework: Programming Design Paradigm, Database Management Systems, Algorithms, NLP, Machine Learning, Mobile App Dev, Software Engineering, Computer Systems
Dayananda Sagar College of Engineering
Bachelor of Engineering in Computer Science · Aug 2018 – Jul 2022
- Graduated with 8.59 GPA
- Key Courses: Operating Systems, Data Structures, Algorithms, Networks, Cloud Computing, AI/ML, Cryptography, Big Data, IoT
Skills & Technologies
The tools and technologies I use to build modern web applications.
Programming Languages
Frameworks & Libraries
Databases & Data
Cloud & DevOps
Work Experience
My professional journey and key contributions.
Associate Software Development Engineer – Backend
Draup · Full-time | Aug 2022 – Nov 2023 · Bengaluru, India (Hybrid)
- Stakeholder for platform modules like Digital Tech Stack, Outsourcing, Customers, and University Page.
- Designed an internal framework for dynamic query generation, enabling real-time aggregation for millions of records and improving chatbot performance by 60%.
- Revamped insight and filter pages with complex Boolean logic modifiers (e.g.,
(A and B) or C), enhancing search expressiveness and boosting engagement by 40%. - Authored 100+ REST APIs using Python & Django REST Framework, integrated Redis caching and Celery for async task handling.
- Implemented subscription-based access control systems and end-to-end data monitoring via Datadog dashboards.
- Migrated APIs from PostgreSQL to Elasticsearch for 5× faster aggregation and near real-time analytics.
- Applied advanced optimization (partitioning, indexing, query restructuring), achieving 400% faster queries and 50% cost reduction.
- Reduced platform downtime from 4% → 1% using proactive alerting via Datadog & AWS CloudWatch.
- Built an internal Django Admin dashboard to monitor failed push jobs from Postgres→Elasticsearch with custom retry support.
Backend Engineering Intern
Draup · Internship | Apr 2022 – Jul 2022 · Bengaluru, India
- Monitored API bugs via Datadog and implemented AWS CloudWatch alarms, reducing incident response time by 30%.
- Improved efficiency of image requests by adding Redis caching, cutting load times by 70%.
- Built and automated database cleanup scripts that enhanced data consistency and query performance by 25%.
Research Assistant (Co-Inventor, Patent Pending)
Dayananda Sagar College of Engineering | Nov 2021 – Sep 2023
- Co-inventor for a patent on Myocardium Wall Motion and Thickness Mapping (App No: 202341086278).
- Developed custom Python algorithms to measure wall thickness from MRI cine scans, increasing precision by 50%.
- Implemented parallelized NumPy and CuPy pipelines for GPU acceleration, reducing runtime by 60×.
- Enhanced the visualization pipeline with fibrosis mapping and multiprocessing using
ThreadPoolExecutor.
Featured Projects
A selection of projects that showcase my skills and interests.
🧠 Real-Time AI Voice Assistant (Twilio + GPT-4o)
Architected a full-stack real-time AI assistant that handles two-way voice calls via Twilio and OpenAI GPT-4o’s streaming API with sub-500ms latency.
📊 Open Jobs Analytics Platform (JobStats)
Built a distributed scraping and analytics pipeline tracking 50K+ postings/day using Celery, Puppeteer, and Prometheus, achieving 99.9% uptime.
🤖 Intelligent Agent System (GPT-4 + Gemini)
Designed a multi-LLM framework using OpenAI GPT-4 and Gemini with FastAPI WebSockets and Discord integration for real-time cross-agent orchestration.
💼 LinkedIn Assist – LLM-Powered Chrome Extension
Developed a Chrome extension that uses GPT-3.5 for real-time job filtering on LinkedIn based on natural-language queries with advanced Boolean logic.
❓ StackOverflow Clone
Full-stack Q&A platform built with React + Node.js (TypeScript), implementing MVC and design patterns like Facade and Strategy with comprehensive Jest/Cypress tests.
🎧 Silent Disco – Multi-Device Audio Streaming
Created a real-time system to stream live desktop audio via WebSockets and Python to multiple Bluetooth devices, enabling synchronized listening sessions.
💬 Aspect-Based Sentiment Analysis using Attention LSTM
Developed an attention-based LSTM for target-dependent sentiment analysis using the SemEval dataset with word embeddings and NER-based aspect extraction.
❤️ Myocardium Wall Motion & Thickness Mapping (Patent-Pending)
Analyzed MRI cine scans to map wall motion and fibrosis regions using NumPy, CuPy, and multiprocessing — performance optimized by 60× via GPU acceleration.
Certifications
My commitment to continuous learning and skill development.
Expert - Programming and Algorithms
CodeSignal
Advanced Retrieval for AI with Chroma
DeepLearning.AI
Supervised ML: Regression & Classification
Stanford / DeepLearning.AI
GitHub Stats
A look at my contribution activity and stats.
Get In Touch
I'm currently open to new opportunities and collaborations.
Let's build something amazing together.