Real challenges. Working solutions.

Green Future uses InspHire as its ERP system. I developed a connected Next.js app that brings machine information, maintenance and planning together, giving internal engineers and the external provider a shared view of each machine.

The starting point was a practical coordination problem: avoiding unnecessary visits by both parties to the same machine. That foundation has since grown into a platform with customer access, machine data and administrative workflows.

Customer access is part of the same platform. Each customer gets a restricted view of their own machines and can report a problem directly from that overview.

Photos and a description give Green Future context for the request. The information arrives together, so the team can assess the issue and organise follow-up.

Not every machine has the required equipment. For those that do, the available measurements are retrieved through APIs and displayed on the map.

More machine data is being collected, but some of it is not yet available through the API. We are working towards predictive maintenance: could the data help anticipate which parts are needed and when to intervene before a breakdown? That predictive capability is still in development; the platform does not currently predict failures.

Previously, PDFs travelled back and forth by email. When a file was too large, the administrator had to remove images for review, then retrieve the original version to send to the customer.

Now the administrator uploads the document for approval, together with additional files of different types. The owner can approve or reject it on a phone and provide a reason for rejection.

After approval, the app creates magic links to the documents that can be pasted into a customer email. Documents remain easy to find in one place, and email attachment limits no longer get in the way.

Claude interprets the customer’s request and converts it into a JSON query. My backend uses that query to retrieve the appropriate customer, items and prices.

The planned integration will display the prepared quote in the existing app. A member of staff will review the selected customer, added items and prices.

Only after human approval should the quote be created in InspHire and a new PDF module generate the document. In-app review, creation in InspHire and PDF generation are still being built.

The pipeline combines Oxford Nanopore sequencing, Dorado and a Python methylation quantification package, allowing researchers to explore multiple methylation types at base-level resolution.