<Abhi Patel>

// full stack developer

// about me

Passionate full-stack developer and systems thinker building scalable, high-performance web applications.

I’m Abhi Patel, specializing in Next.js and TailwindCSS on the frontend and Node.js on the backend. I design reliable infrastructure using Docker, Kubernetes, and Google Cloud.

Beyond web development, I explore machine learning and AI, experimenting with PyTorch, Python, and large language models to create intelligent systems and procedural tools.

I thrive on understanding systems from the ground up—whether it’s optimizing backend performance, designing distributed systems, or crafting seamless user experiences.

Outside of coding, I pursue photography, capturing cinematic and atmospheric visuals that inspire both my creative and technical work. My projects reflect a balance of curiosity, technical rigor, and craftsmanship, and I’m always eager to tackle challenges that push my skills further.

// skills

// work experience

january 2025 - present

Panda Hub Inc, ON

// full stack developer

  • . 60% improvement in operational efficiency by migrating the web app and admin dashboard from Flutter to Next.js, refactoring legacy code and adding new features.
  • . Documented and analyzed undocumented backend code, creating internal guides that reduced onboarding time and improved team maintainability.
  • . Integrated Google Pay, Apple Pay, and Stripe with a custom UI, enhancing checkout experience and increasing successful transactions.
  • . Built an automated rescheduler system that cut appointment cancellations by 20% and boosted customer satisfaction.
  • . Migrated marketing website to Next.js with a Directus CMS pipeline, improving Largest Contentful Paint (LCP) and load times by 90%, and optimized dynamic components using Cloudflare edge services.
may 2024 - september 2024

Loyalist College in Toronto, ON

// full stack developer, internship

  • . Led the team of 6 to develop a fully automated web-based mock interview tool, utilizing cutting-edge machine learning and web technologies to enhance interview preparation for users.
  • . Developed a facial expression classification model achieving 77% accuracy by fine-tuning MobileNet on the Facial Expression Recognition Dataset.
  • . Designed and implemented frontend features, including background presence detection using an object detection model to detect other persons and background noise levels detection, despite challenges with model accuracy.
  • . Identified and resolved system bottlenecks, leading a team-driven effort to reduce runtime by 70%, enhancing tool efficiency and user experience.
  • . Migrated marketing website to Next.js with a Directus CMS pipeline, improving Largest Contentful Paint (LCP) and load times by 90%, and optimized dynamic components using Cloudflare edge services.
june 2021 - march 2024

Codise, CA

// full stack developer

  • . Architected and implemented a full-stack solution using TypeScript, React, Node.js, and PostgreSQL, enabling the company to perform daily operations online, streamlining processes and reducing manual workload by 50%.
  • . Created a robust scheduling system that automates job assignment and workflow stages, reducing manual scheduling errors by 60%.
  • . Designed and developed an end-to-end management module, optimizing resource allocation and ensuring seamless inter-departmental workflow, leading to a 30% increase in operational efficiency.
  • . Implemented comprehensive database logging and tracking, ensuring all stages of the printing process are accurately monitored and traceable.
december 2020 - december 2022

BosLeo Pvt Ltd, GJ

// frontend developer

  • . Developed and maintained the frontend of a Health Infomatrics System using Angular 12, enhancing user experience for over 500+ staff.
  • . Integrated Laravel-based backend APIs to ensure seamless data flow and improve system reliability by 30%.
  • . Boosted UI performance by refactoring and optimizing Angular table components, resulting in a 30% reduction in load times across the project.
  • . Created a global table indexing component that dynamically adjusted based on pagination, improving data navigation and user interaction.
  • . Designed and implemented reusable Angular components for radiology and pathology report comparisons, improving code maintainability and consistency.
  • . Resolved numerous bugs, significantly enhancing system stability.

// projects

Tic Tac Toe Multiplayer (Microservices)

Real-time multiplayer game built on a distributed microservices stack, coordinating users, game state, and matchmaking via scalable APIs and event-driven messaging.

What I learned: Architecting stateless services, handling distributed state, and orchestrating real-time communication across service boundaries.

Rust + WASM + HTML5 Text Editor

In-browser text editor leveraging Rust compiled to WebAssembly for fast, memory-safe editing and advanced text manipulation directly in the browser.

What I learned: Integrating Rust with JavaScript, optimizing for WASM performance, and bridging low-level system code with web UI.

Coffee POS Microservices Web App

Modular point-of-sale system for coffee shops, using containerized microservices for orders, inventory, and analytics, supporting real-time updates and robust fault tolerance.

What I learned: Building resilient business workflows, implementing service discovery, and ensuring transactional consistency across distributed systems.

Bento Grid + Image Generator

Math-driven bento grid generator using CSS Grid, allowing users to define rows, columns, and merge cells to design custom layouts. The layout is converted into a reference image and combined with a prompt to generate visually cohesive images per cell using Nano Banana Pro 🍌

What I learned: Translating abstract layout logic into visual systems, generating and using reference images for multimodal AI models, and blending algorithmic UI design with generative image workflows.