Skip to content

hi, I'm

Riddhi
Patel

Backend Software Engineer focused on developer tools, distributed systems, and AI-powered applications.

01 · About

About me

I'm a backend software engineer based in Mumbai who enjoys building reliable systems and solving the tricky problems behind the scenes.

I worked on BrowserStack's SDK team from January to July 2026. Across five languages (C#, Ruby, Node.js, Python, and Java), I fixed session lifecycle, capability propagation, and session-linking issues, and improved SDK reliability.

Previously, I was a backend developer at LegalGini, an AI legal platform built on Django. I built REST APIs, reduced measured document-processing time by approximately 69% (8 minutes to 2.5 minutes), and used Redis and Celery to run slow tasks in the background.

I care about writing clean code, testing it properly, and making sure things keep working when something goes wrong. Lately, I've been exploring distributed systems, AI infrastructure, and scalable backend architectures.

~69%

document-processing time reduction at LegalGini (8 min → 2.5 min)

~180

framework tests validated at BrowserStack

5

languages used in production

8.77 / 10

B.E. Information Technology, University of Mumbai

Education

B.E. Information Technology

University of Mumbai

CGPA 8.77 / 10

02 · Experience

Where I've shipped

  1. BrowserStack

    Jan 2026 to Jul 2026

    Backend Intern (SDK Team) · Mumbai, India

    Worked on session lifecycle, capability propagation, and session linking across C#, Ruby, Node.js, Python, and Java SDKs.

    Built FileLock-based leader/follower coordination for distributed Python workers, fixing Automate-to-Observability session linking.
    Implemented Reqnroll and MSTest flows across SDK, gRPC binary, and App Accessibility layers, validated across approximately 180 framework tests.
    Enhanced framework-detection and SDK-adoption telemetry feeding BigQuery analytics.
    Led root-cause analysis across 7+ repositories, shipping coordinated fixes and strengthening regression coverage.
    C#
    Ruby
    Node.js
    Python
    Java
  2. LegalGini

    Jun 2025 to Oct 2025

    Software Developer Intern · Remote

    Built Django backend services and document-processing workflows for an AI legal platform.

    Developed REST APIs and Docker/Docker Compose setups for local development and containerized deployment.
    Reduced measured document-processing time by approximately 69%, from 8 minutes to 2.5 minutes, through concurrent Textract OCR, improved Gemini batching, and overlapping OCR with extraction.
    Ran document processing as Celery background tasks, using Redis as the broker/result backend and thread pools for concurrent calls within each task.
    Built semantic document retrieval and chat workflows using chunked OCR text, OpenAI embeddings, and Pinecone.
    Python
    Django
    Redis
    Celery
    Docker
    OpenAI
    Gemini
    Pinecone
    LangChain
  3. Earlier internships

    • Vault of Codes · Python Developer Intern, Built Python applications and automation scripts.
    • Coding Raja Technologies · Web Development Intern, Developed responsive web interfaces and full-stack features.

03 · Skills

Tools I reach for

Backend Engineering

Databases

Distributed Systems

AI & LLM

  • OpenAI API
  • Gemini API
  • LangChain
  • Pinecone
  • RAG Pipelines
  • OCR / AWS Textract

DevOps

The linked ones go straight to the code. Have a look.

04 · Projects

Things I've built

YogaBliss: Wellness Platform

A wellness application with user authentication and activity tracking. Built a React frontend backed by Node.js/Express REST APIs and MongoDB, implementing JWT authentication, activity dashboards, and schemas for user and activity data.

Result: Connected authentication and activity tracking across the frontend and backend.

  • Node.js
  • Express
  • MongoDB
  • React
  • JWT

YogaBliss: Pose Detection

A Flutter application that identifies yoga poses from camera input. Trained a CNN classifier and deployed it on-device using TensorFlow Lite for real-time pose detection.

Result: CNN classifier achieved 85% test accuracy across five yoga poses.

  • Flutter
  • Dart
  • TensorFlow Lite
  • CNN

HydraSip: Freelance Landing Page

A client website presenting premium bottled water and custom-label services. Designed and built the landing page in Next.js and TypeScript, then deployed it on Vercel.

Result: Delivered a production website to a real client.

  • Next.js
  • TypeScript
  • Vercel

Resume Builder

A browser-based tool for creating resumes with live preview and PDF export. Built resume editing and export functionality using HTML, CSS, and JavaScript, with no backend required.

Result: Users can create, preview, and export a resume entirely in the browser.

  • JavaScript
  • HTML
  • CSS
  • PDF Generation

05 · Leadership

Beyond the code

Publicity & PR Head

VCET Hackathon Committee

Led outreach for the college hackathon across 20+ campuses.

Joint Secretary

VCET Core Coding Club

Organized coding contests and technical workshops.

Organizing Head

VCET Placement & Training Cell

Coordinated placement drives and training programs.

Organizing Head

VCET Women Empowerment Team

Ran events supporting women in engineering.

06 · Off the record

Some honest numbers

100+

flaky tests investigated

4

college leadership roles

5

production languages worked with

07 · Contact

Let's talk

I'm currently looking for Backend Software Engineer and Software Engineer opportunities, and I'm always happy to talk about distributed systems, automation, or AI. Email is the quickest way to reach me, and I usually reply within a day.