Snap a photo. Get a full HSE report.
Safety reporting in the field is meant to be quick, but long forms and clunky apps mean hazards go unlogged. So we built Fore-Site — our own AI-powered progressive web app that turns a single photo into a complete observation, incident or hazard report. The AI reads the image, drafts the details and the worker just checks and submits. It works offline, syncs when reconnected, and feeds a desktop command centre for tracking, investigation and approvals.
From a photo in the field to a signed-off report.
The report that never gets
written is the one that matters.
On a busy site, spotting a hazard is the easy part. Logging it is where reporting falls down: the worker has to stop, open an app, and fight through long forms — picking categories, rating risk, writing a description — often with gloves on, in poor signal, with a dozen other jobs waiting. So the near-miss goes unreported, and the paper trail that keeps people safe never gets started.
The teams that need this most — construction, energy, manufacturing, industrial — work in exactly the conditions that make traditional reporting hardest: no desk, patchy connectivity, and no time. And even when reports do land, safety managers are left stitching them together by hand to see what's actually happening across their sites.
Four pillars behind Fore-Site.
Fore-Site is our own product — an AI-assisted reporting app paired with a desktop command centre. We designed and built it around how field teams actually work: fast to capture, honest when offline, and complete by the time it reaches a manager. Four pillars carry it.
AI-assisted capture
Point the camera at a hazard and the AI reads the scene — it classifies the hazard, drafts a description, sets a risk level and priority, and pre-fills the report with a confidence score. The worker checks, edits if needed, and submits in seconds.
Offline-first PWA
Built as a progressive web app, so it installs like a native app with no app store and works with no signal. Reports are captured and queued on remote or underground sites, then sync automatically the moment the device reconnects.
Desktop command centre
Every report flows into a web command centre where safety teams triage, investigate and approve. Multi-stage approvals, CAPA, inspections, a hazard library and AI root-cause analysis turn raw reports into tracked, closed-out actions.
Configurable for any site
Sites, locations, modules, custom fields and templates are all configurable, so one platform fits construction, energy, manufacturing and industrial operations without a rebuild — governed, auditable and ready for compliance.
What Fore-Site changes.
Our own product,
live and evolving.
Fore-Site launched as a V1 MVP in May 2026 — a working AI reporting app and command centre, built entirely in-house. It's the product we point to when we talk about applied AI: real image analysis doing real work in the hands of field teams.
It was built to grow. New modules, hazard types and AI capabilities slot in without unpicking what's already there, and enhancements are shipping continuously.
A full observation, incident or hazard report from a single photo
Reporting time cut from minutes of form-filling to seconds
Consistent hazard classification and risk scoring across every site
Works offline and syncs on reconnect — nothing lost in the field
A live command centre for triage, investigation and approvals
One configurable platform across construction, energy and industrial sites
Where AI does the work in Fore-Site.
Four capabilities carry AI from the moment a photo is taken through to the pattern a safety team acts on — not a chatbot bolted on top, but the thing that makes "seconds instead of minutes" actually true.
From a photo to a filed report — no typing required.
Computer vision is what makes Fore-Site's core promise possible: point a phone at a hazard and get back a classified, risk-scored, ready-to-submit report. Then AI root-cause analysis in the command centre turns individual reports into a pattern safety teams can actually act on, instead of a pile of PDFs nobody has time to cross-reference.
A single photo becomes a full report
Point a phone at a hazard and computer vision reads the scene — classifying what it's looking at and drafting a plain-language description before the worker has typed a word.
Risk scored the moment it's captured
Every report gets an automatic risk level, priority and confidence score straight from the image — so the most serious hazards surface first, not whichever one happened to get logged first.
Root-cause analysis, not just a report
The command centre runs AI analysis across incoming reports to help safety teams see the pattern behind a hazard, not just the individual occurrence — turning raw reports into a trend a team can act on.
Consistent classification, every site, every time
The same hazard reads the same way whether it's logged on a construction site or in a warehouse — removing the inconsistency that comes from different people describing things differently by hand.
Shipped as an MVP, and evolving.
Version 1 — MVP
May 2026
Ongoing enhancements
Currently in progress
What Fore-Site is built with.
A React progressive web app on the front line, a Python and PostgreSQL backend on AWS behind it — proven pieces, chosen to hold up on real sites.
AWS
Python
PostgreSQL
React.js
PWAOther products we've shipped.
Real-time carbon & compliance reporting for cold chain.
A data management platform built to manage ESG data and deliver ESG insights, analytics and reports to the business and its stakeholders.
Time-critical incident reporting, built for the field.
A realtime incident reporting application built to collect, escalate and report on incident data across the cold chain and construction business.
A single source of truth for prospect intelligence.
A centralised platform to manage the acquisition, cleansing, matching and enrichment of prospect data at scale — replacing fragmented sources and manual scoring with one unified system.
A stalled platform, rebuilt into an award-winning product.
An award-winning data management and lead generation platform, capable of tracking and reporting campaign data in real time.
A unified analytics platform from scattered, siloed data.
Built to aid sustainable economic development, Hanse helps businesses identify crucial international trade patterns — winner of Innovate UK funding.
Validating B2B leads against live LinkedIn data, automatically.
A tool that validates whether a lead still works at a company, with real-time lookups against their LinkedIn profile.
Questions, answered straight.
Have an outcome in mind?
We'll help you
ship it.
- You're building a data product and need a team that can deliver.
- You want to get AI-ready — pragmatically, not theoretically.
- Your reporting is a mess and you need a real platform underneath it.