Paper job cards and route cards
Works orders, route cards and inspection sheets travel with the job on paper, so progress is only known when someone walks the floor.
Manufacturing & engineering
AI & Software Development for UK Manufacturers
We help manufacturers, engineering firms and food producers replace paper job cards and spreadsheet schedules with shop-floor software that feeds your ERP — and add AI where it genuinely helps, from visual inspection to reading supplier certificates.
- B2B ordering and IoT dashboard projects
- Works alongside your ERP, not against it
- Inspectors confirm every AI finding
Industry challenges
Where manufacturers lose hours between systems.
The machines are usually fine. The time goes on paperwork, re-keying and finding out what is actually happening on the floor.
Sensible first steps
- Tablet job tracking that updates the ERP as work moves
- Batch traceability from goods-in to dispatch
- A trade portal for repeat orders and order status
- Maintenance logs with reminders and parts history
Scheduling in a spreadsheet
The production plan lives in a spreadsheet one person maintains, disconnected from stock, orders and machine availability in the ERP.
Traceability that takes days
Linking a finished batch back to material certificates, operators and inspection results means digging through folders when a customer or auditor asks.
Maintenance kept in people’s heads
Planned maintenance, breakdowns, spare parts and PUWER or LOLER inspection records are scattered across notebooks, emails and one engineer’s memory.
Trade orders by email and phone
Customers email PDF purchase orders or phone in repeat orders, and staff re-key them into the ERP, then chase stock and delivery dates by hand.
Machine data nobody looks at
Sensors and controllers already record running time, temperatures and faults, but the data never reaches a screen anyone uses to make decisions.
Solutions
What we build for manufacturers.
Practical software for SME and mid-market plants — shop-floor tools, quality records, customer portals and data dashboards — connected to the ERP and accounts systems you already run.
Shop-floor job tracking
Tablet and scanner screens to start, pause and complete operations, record scrap and rework, and show work in progress — a lighter alternative to a full MES.
Learn more
Quality and traceability records
Inspection checklists, non-conformance reports, corrective actions and batch genealogy linked to material certificates — ready when a customer or auditor asks.
Learn more
ERP and system integration
Shop-floor tools, portals and warehouse scanning connected to your ERP and accounts package through their APIs or data exchange, so orders, stock and jobs stay in step.
Learn more
B2B ordering and export portals
Trade portals with account pricing, repeat orders, order status and invoices, plus export paperwork data such as commodity codes and Incoterms taken from each order.
Learn more
Machine and sensor dashboards
Running time, stoppages, temperatures and energy use from connected machines in live dashboards — working with your controls engineer on the machine side.
Learn more
Maintenance management
Planned maintenance schedules, breakdown logging, spare parts and PUWER and LOLER inspection records, with reminders and a history for every asset.
Learn more
AI use cases
AI on the shop floor and in the office.
AI that looks, reads and predicts — tested on your own images, documents and machine data first, with operators, inspectors and planners making the final call.
Visual defect detection
Computer vision checks parts or packs for scratches, missing features, labels and fill levels, and sends doubtful items to an inspector. We start with a feasibility study on your images.
Predictive maintenance
Spot unusual vibration, temperature or current patterns in sensor data and raise a maintenance check before a likely failure — once there is enough history to learn from.
Certificate and supplier document reading
Extract grades, batch numbers and test results from certificates of conformity and material certificates, check them against the purchase order and flag mismatches.
Purchase order capture
Read customer POs from emails and PDFs, match part numbers and prices, and create draft sales orders in your ERP for staff to confirm.
Demand forecasting
Forecast demand by product and customer from order history and seasonality, to plan raw materials, shifts and stock ahead of busy periods.
Work instruction assistant
Answer questions from your SOPs, manuals and past fixes with a link to the source page, so new starters find answers without hunting through folders.
Technology & integration
Technology that connects the floor to the office.
We are software engineers, not a controls integrator. We take machine data from gateways and APIs and work alongside whoever looks after your PLCs — and we do not supply hardware.
Machine and sensor data
MQTT and IoT gatewaysAWS IoTTime-series storageAlerts and thresholds
Business systems
ERP APIs and data exchangeAccounts packages such as Xero or SageEDI and CSV importsLabel and barcode printing
Shop-floor devices
Tablets and touchscreensBarcode and QR scanningOffline-capable web appsLarge-screen status boards
Applications
React and Next.jsNode.js, Python and C#/.NETFlutter and React NativePostgreSQL
AI and vision
PyTorch and TensorFlowImage classification and detectionAnomaly detectionDocument extraction and OCR
Hosting
AWS, including UK regionsOn-site servers where neededDockerBackups and monitoring
Security & data
How we secure production and design data.
Manufacturing systems hold drawings, recipes, pricing and customer data, and connect to equipment on the floor. These are our default practices, not a compliance badge.
Machine networks kept separate
Shop-floor connections are designed with your IT team to keep machine networks apart from office and internet traffic, with one-way data flows where possible.
Access by role and site
Operators, supervisors, quality, engineering and customers each see only what they need, and shared logins are avoided wherever individual records matter.
Drawings and recipes protected
Commercially sensitive files are encrypted and access-logged, and customer drawings are handled under the confidentiality terms you agreed with each customer.
Tamper-evident quality records
Inspection results, sign-offs and corrections are logged with who, when and why, rather than silently overwritten.
People confirm AI findings
Vision and prediction models flag and suggest; inspectors and engineers make the accept, reject and maintenance decisions.
Works through network drops
Shop-floor apps queue data on the device and sync when the connection returns, so a network drop does not stop production records being captured.
We are a software development company, not a quality consultancy, certification body or controls integrator. We build to the requirements your quality, health and safety and customer standards set — for example ISO 9001 or customer approvals you hold — and provide the technical evidence your auditors ask for. Sign-off stays with your organisation.
Relevant work
Products we have designed and built.
Related projects from our case studies: a B2B catalogue, bulk ordering and export platform for a manufacturer, and a connected-device dashboard for schedules and energy use.
All case studiesHow we work
How a manufacturing project fits around production.
Seven stages, the same on every project. The length of each one changes with the work; skipping one never saves time for long.
- 01
Discovery
We walk the floor with supervisors, operators and quality staff, follow a job from order to dispatch, and list every paper form, spreadsheet and system it passes through.
Current-state process map
- 02
Strategy
We agree what to build first and what to leave out, pick the engagement model, and set milestones with a budget range in pounds.
Scoped first release and estimate
- 03
UX & architecture
Screens are designed for the shop floor — large touch targets, gloves, noise and poor lighting — and tested with operators on the devices they will use.
Operator-tested prototype
- 04
Development
Two-week sprints with a demo at the end of each, built against a test copy of your ERP rather than the live system until everyone is confident.
Working features against a test ERP
- 05
Testing
Automated tests written as we build, then manual QA, accessibility and performance checks before anything goes live.
Test evidence and a release checklist
- 06
Launch
We pilot on one line, cell or product family, run alongside the paper process until the data matches, then roll out in stages planned around your shutdowns and busy periods.
Pilot sign-off and rollout plan
- 07
Optimisation & support
After launch we fix, measure and improve — a support retainer, a roadmap of next features, or a clean handover to your own team.
Support plan or handover
FAQs
Questions we get asked.
Straight answers on scope, cost, timelines and how we work. If yours is not here, ask us directly.
Do we need a full MES, or is something lighter enough?
Many SME manufacturers need job tracking, quality records and live work-in-progress rather than a full MES. We usually start there, connected to your ERP, and extend it as needs grow. If a packaged MES would suit you better, we will say so during discovery.
Can you integrate with our ERP?
Usually, through its API, database views or scheduled imports and exports, depending on what your ERP and its licence allow. We confirm the options with you and your ERP supplier in discovery, and test against a copy before touching live data.
Can AI really spot defects on our line?
Sometimes, and it is worth finding out cheaply first. Results depend on lighting, camera position, part variety and how many examples of real defects you have. We run a feasibility study on your own images before any production build, and an inspector always confirms what the model flags.
Is predictive maintenance realistic for a smaller plant?
It can be, if your critical machines have sensors or can be fitted with them, and you have some history of failures. Many plants start with condition monitoring and alerts, which are useful on their own, and add prediction once there is enough data to learn from.
Do you supply sensors, PLCs or other hardware?
No. We are a software company. We work with your controls engineer, machine builder or systems integrator, who looks after the hardware and PLCs, and we bring the data from gateways and APIs into software your team actually uses.
Can you help with traceability for audits and recalls?
Yes. We link batches to material certificates, operators, machines and inspection results, so tracing forwards or backwards becomes a search rather than a hunt through folders. The records support your quality system — ISO 9001, BRCGS or a customer’s requirements — while your quality manager stays responsible for it.
Can you build a B2B portal for our trade customers?
Yes. We build trade portals with account-specific pricing, repeat ordering, order tracking and invoice history, linked to your ERP for stock and prices. For exporters, we can generate the data your shipping and customs paperwork needs from each order.
How much does manufacturing software cost?
It depends on the processes, integrations and number of sites. A discovery sprint starts from £2,000 and gives you a scoped plan and estimate; builds are then fixed price with milestone billing. For a longer programme of work, a managed team starts from £9,500 a month.
Start a project
Planning a manufacturing software or AI project?
Tell us what you make, how work moves across the floor and which systems you run. An engineer will reply within one working day with questions and a practical first step — often a small pilot on one line.
What happens next
- 1A senior engineer reads your brief within one working day, and replies with questions or a first view.
- 2A 30-minute call to understand the goal, constraints and what good looks like — no sales script.
- 3A written proposal with scope, milestones, team and a GBP estimate you can take to your board.
Techsleight Labs is a trading name of Krapton IT Consultancy.
Explore