Full-Stack Software Developer Nov 2022 – Oct 2023

Lisssen & Co

Case study of full-stack engineering at Lisssen & Co, focusing on custom Web Audio playback optimization, MERN architecture, and self-managed Linux/AWS Lightsail infrastructure.

Company Context & Mandate

Lisssen & Co was a digital audio platform and creator ecosystem built to stream high-fidelity audio, connect listeners with creators, and provide audio distribution tools.

I served as a Full-Stack Software Developer, responsible for core platform engineering. My mandate covered three main pillars:

  1. Building a responsive custom audio player engine capable of seamless, low-latency playback, waveform visualization, and buffer management across desktop and mobile browsers.
  2. Developing full-stack feature APIs and database models using the MERN (MongoDB, Express, React, Node.js) stack.
  3. Provisioning, securing, and maintaining the platform's production server infrastructure on Linux (Ubuntu) and AWS (Lightsail & S3) without a dedicated DevOps team.

What I Built

1. Web Audio Engine & Custom Playback Optimization

Engineered a custom audio player abstraction layer on top of HTML5 <audio> and the Web Audio API, paired with Redux state orchestration:

  • Adaptive Chunk Preloading & Range Requests: Implemented byte-range HTTP request handling (Range: bytes=...) from server to client, allowing instantaneous playback initiation without waiting for full audio files to download.
  • Buffer Strategy & Re-buffering Protection: Designed a prefetching algorithm in Redux to buffer upcoming queue tracks in memory, eliminating gaps between track transitions.
  • Waveform & Seeking Performance: Abstracted audio scrubbing into a non-blocking UI task using requestAnimationFrame, preventing frame drops during rapid playback seeking.
[ User Action ] ──> ( Redux Queue Manager ) ──> ( HTML5 / Web Audio API )
                                                        │
                                                        ▼
[ AWS S3 Storage ] <── ( Signed Range Requests ) <── ( Express API Proxy )

2. MERN Stack Feature Architecture

  • MongoDB & Mongoose Schemas: Modeled complex audio metadata, user playlists, listening histories, and artist permissions with indexing strategies for fast query execution.
  • Express.js REST APIs: Authored endpoints for file upload validation, user authentication (JWT), track streaming metadata, and social interaction handling.
  • React & Redux UI: Built dynamic dashboards for listeners and creators, leveraging Redux Toolkit to maintain centralized player state across all site routes.

3. Self-Managed Linux & AWS Infrastructure

  • AWS Lightsail & Nginx: Configured and maintained Ubuntu Linux server instances on AWS Lightsail. Built Nginx reverse proxy configurations with SSL termination (Let's Encrypt), Gzip/Brotli compression, and rate-limiting rules.
  • Process & Health Management: Used PM2 and Systemd services for continuous process monitoring, environment variable isolation, and automated service auto-restarts.
  • AWS S3 & Media Security: Designed secure asset storage where private audio files were stored on S3 and served exclusively via short-lived signed URLs generated by the Node.js backend.

Scope & Responsibilities

  • End-to-End Application Ownership

    Owned features from database schema design down to UI components and server deployment.

  • Infrastructure Administration

    Managed system security, firewall rules (UFW), SSL certificates, database backups, and server performance monitoring.

  • Client & Mobile Optimization

    Ensured low data footprint and cross-browser compatibility for mobile web listeners.

Impact & Results

Playback Latency

Reduced initial audio playback start time to under ~300ms via HTTP Range requests and predictive preloading.

Infrastructure Uptime

Maintained high platform availability across Linux VPS instances throughout deployment cycles.

Bandwidth Optimization

Reduced origin server egress costs by offloading static assets and audio byte-streams directly to S3 storage via signed URLs.

Tech Stack Deep Dive

React & Redux
Interactive user interface rendering and centralized audio player state management.
Node.js & Express
Scalable event-driven backend API framework handling stream requests and business logic.
MongoDB & Mongoose
NoSQL document database for audio track metadata, user profiles, and playlist models.
Linux (Ubuntu)
Base server operating system managed via CLI, systemd, and custom shell scripts.
AWS Lightsail & S3
Cloud compute hosting and scalable object storage for media assets.

Links & Resources

Lessons Learned

  • 01

    Audio State Requires Strict Isolation

    Media player state must be decoupled from standard UI component lifecycles to prevent audio interrupts when navigating web routes.

  • 02

    Infrastructure Ownership Builds Engineering Depth

    Managing Linux servers and cloud storage directly instills a deep appreciation for network latency, security, and resource allocation.

  • 03

    What I'd Do Differently

    I would introduce a dedicated CloudFront CDN distribution in front of the S3 media buckets to further decrease playback latency for international listeners.

More from Experience