TH
HomeAboutProjectsClientsBlogContact
ENID
ENID
Taufik Hidayat

Full Stack Developer building modern web experiences.

Navigation

  • Home
  • About
  • Projects
  • Clients
  • Blog
  • Contact

Connect

© 2026 Taufik Hidayat. All rights reserved.


  1. Home
  2. Projects
  3. EVERCAM
🇬🇧 United KingdomAutomotiveMarch 2024

EVERCAM

A construction site monitoring platform with AI-powered camera intelligence.

EVERCAM screenshot 1
Fullscreen

About this project

Evercam is a UK-based SaaS platform that provides 24/7 construction site monitoring through a network of time-lapse and live-stream cameras. The platform aggregates footage from thousands of cameras across hundreds of active construction sites, enabling project managers, contractors, and stakeholders to monitor progress remotely, resolve disputes with recorded evidence, and automatically generate time-lapse progress videos.

500+
Sites Monitored
2,000+
Cameras Online
12
Countries
99.5%
Uptime SLA

Key Responsibilities

  • 1Developed and maintained key frontend modules of the main dashboard using Next.js and TypeScript.
  • 2Built an interactive camera map view with real-time status indicators and live preview thumbnails.
  • 3Implemented time-lapse video generation UI with clip selection, speed controls, and export options.
  • 4Integrated with Evercam's REST API and WebSocket feeds for live camera status and alert notifications.
  • 5Collaborated with the backend team on API contract design for new features using OpenAPI specs.
  • 6Improved dashboard load performance by refactoring large component trees and introducing pagination.

Challenges

Rendering thousands of live camera feeds efficiently

Loading 2,000+ camera thumbnails simultaneously caused browser memory exhaustion. Implemented a virtualised grid with Intersection Observer-based lazy loading and a thumbnail refresh interval that backs off for off-screen cameras.

Real-time status sync across unreliable camera connections

Cameras on construction sites frequently drop connectivity. Designed a hybrid polling + WebSocket approach: WebSocket for instant updates when available, with a 30-second polling fallback and visual indicators for degraded connections.

Accomplishments

  • Reduced dashboard initial load time by 55% through virtualised rendering and smarter API pagination.
  • Delivered a new time-lapse editor feature that became one of the top-requested capabilities by clients.
  • Contributed fixes and improvements that helped maintain the platform's 99.5% uptime SLA.
Date
March 2024
Country
🇬🇧United Kingdom
Sector
Automotive

Tech Stack

Next.jsOpenAI APIVercel AI SDKSupabaseTailwind CSS
Previous
Smartgenix
Next
Klarsmile