HomeServices
What we do

Six practices. One accountable team.

From the first architecture review to a production incident at 2am, the same people stay on your account. No handoffs between departments, no re-explaining your business to a new team every quarter.

6Service practices
40+Systems delivered
2 wksTo your first working demo
SpecializationsSoftware ConsultingCustom DevelopmentCloud & InfrastructureData & AnalyticsDigital TransformationManaged IT & Support

Not sure which practice you need?

Most people arrive with a symptom rather than a solution. Pick whichever sounds closest to your situation.

Recommended practice

Read more about it
Specialization

Shop-floor systems: our focus area

Within Custom Development and Digital Transformation, this is where we spend most of our engineering time. Most factories already have the data — it just sits in four registers that never agree with each other. We build the system that joins them, designed for people wearing gloves rather than office workers.

What we hear on the floor
The supervisor fills the production register at the end of the shift, mostly from memory.
Stock says we have material. The store says we ran out on Tuesday.
We only find out about a quality problem when the customer calls.
Attendance is in a separate book, so we cannot tell which shift is actually more productive.
Four modules, one shared record
MES

Production Tracking

Manufacturing Execution System

Replaces the paper production log. Operators record output, downtime, and scrap from a tablet at the line, and the plant manager sees it while the shift is still running rather than the next morning.

  • Output logged per line, per shift
  • Downtime with reason codes
  • Scrap and rework tracking
  • Live target vs actual
QMS

Quality Control

Quality Management System

Inspection checklists that must be completed before a batch moves on, with defects logged against the batch, the shift, and the machine — so patterns surface before the customer finds them.

  • Digital inspection checklists
  • Defect logging with photos
  • Batch traceability
  • Corrective action tracking
WMS

Inventory & Materials

Warehouse Management System

Raw material in, material issued to each line, finished goods out. Consumption is deducted automatically when production is logged, so stock stays close to accurate without a manual count.

  • Goods receipt and issue
  • Auto-deduction on production
  • Reorder level alerts
  • Stock movement history
WFM

Attendance & Shifts

Workforce Management

Clock-in via badge scan or PIN on a shared tablet — no biometric hardware needed to start. Because attendance and production share a record, you can finally compare shift against shift on real output.

  • Badge or PIN clock-in
  • Shift rosters and overtime
  • Output per shift comparison
  • Absence and late reporting

Modules can be adopted one at a time. Most factories start with production and attendance.

How we do it

Five stages from first floor visit to retiring the paper register. It plays automatically — pause it or step through at your own pace.

Built for the floor

The most common reason factory software fails is not the software. It is that the person on the line quietly goes back to the paper register because the screen was designed for an office.

  • Works offline. Factory wifi drops. The tablet queues entries locally and syncs when it reconnects — the operator never notices.
  • Large touch targets. Designed for someone in gloves glancing at a screen for four seconds.
  • One decision per screen. Log output. Flag a defect. Request material. No nested menus.
  • Local language on the floor. Worker screens in Hindi, Gujarati, or your regional language; the office stays in English.
  • Shared tablet, not personal phones. Badge scan or PIN identifies who is logging. No app install for anyone.
Shop-floor first
Offline-capableOne pilot line to start
How customization works

Every factory is different, but not every difference needs custom code. We use three tiers, and we tell you honestly which one your requirement actually needs.

01

Configuration

Your product names, process stages, shift patterns, checklist items, and language — changed through settings. No code, no deployment.

Included
02

Automation rules

Alert the plant head if scrap crosses 5%. Block a batch until inspection is signed. Notify the store below reorder level. Built on a shared rule engine.

Small add-on
03

Bespoke module

Genuinely unique needs — an ERP or Tally integration, a machine data feed. Built as a plug-in to the same platform, so you keep one login and one database.

Quoted per scope

We will tell you when tier 1 solves what you thought needed tier 3. It is cheaper for you and easier for us to support.

How we roll it out
Weeks 1–2

Map and configure

We walk your floor, watch a full shift, and configure the platform to your actual stages, products, and shift patterns. No development yet — just fitting to how you already work.

Weeks 3–6

Pilot on one line

Production and attendance go live on a single line, running in parallel with the paper register. We sit with operators during the first shifts and fix what annoys them immediately.

Weeks 7+

Extend and retire paper

Once the system and the register agree, the register goes. Then inventory and quality are added, and we roll to remaining lines with a team that has already used it.

Is this right for your factory?

A good fit if

  • Your shop floor still runs on paper registers or spreadsheets
  • You have roughly 20 to 500 people on the floor
  • Departments report numbers that do not reconcile
  • You want to start small and prove it before committing
  • Your process does not fit any off-the-shelf product cleanly

Probably not, if

  • You already run a working MES and want only minor changes
  • Deep PLC and SCADA machine integration is the first requirement
  • You want full automation with no human data entry at all
  • You need it live across every line in under four weeks
  • An off-the-shelf product already fits your process — use it
Book a floor visit
01

Software Consulting

Before anyone writes code, we work out whether code is even the answer. We audit what you already run, interview the people who use it daily, and give you a straight recommendation — including when that recommendation is to buy something off the shelf, fix a process, or do nothing yet.

What we hear before we start
We have six systems and none of them talk to each other. Nobody can tell us which one is the source of truth.
Our last vendor quoted eighteen months. Is that normal, or are we being taken for a ride?
We know something is wrong with our stack but we do not have anyone senior enough internally to say what.
What you get
  • Technical audit of existing systems
  • Stakeholder interview findings
  • Build vs. buy cost comparison
  • Sequenced technology roadmap
  • Risk register with mitigations
  • Vendor and platform shortlist
Typical tools
Architecture reviewCost modellingProcess mappingVendor assessment
Discuss this practice
2–6 weeks
AuditRoadmapRecommendation
Typical duration
2–6 weeks
Team size
1–2 senior consultants
Commercial model
Fixed fee
You keep
Written findings + roadmap
02

Custom Development

Web and mobile applications built around your actual workflow rather than bent to fit someone else's product. We ship working software every two weeks, so you watch it take shape instead of waiting months for a reveal — and you can change direction while changing direction is still cheap.

Our deepest focus within this practice is shop-floor systems — MES, QMS, WMS, and WFM — built as a modular platform rather than four disconnected tools.

What we hear before we start
The off-the-shelf tool does 70% of what we need and we are doing the other 30% in spreadsheets.
Our current system was built by someone who left. Nobody knows how it works.
We need this live before the season starts and every quote we have had says after.
What you get
  • Working software every two weeks
  • Full source code and repository access
  • Automated test suite
  • CI/CD pipeline
  • Technical documentation
  • Handover and training sessions
Typical tools
ReactNext.jsTypeScriptNode.jsPythonReact NativePostgreSQLMESQMSWMSWFM
Discuss this practice
8–24 weeks
FrontendServicesDatabase
Typical duration
8–24 weeks
Team size
2–5 engineers
Commercial model
Fixed scope or pod
First demo
Within 2 weeks of kickoff
03

Cloud & Infrastructure

Migration and architecture across AWS, Azure, and GCP — sized for the load you actually have rather than the one on the slide. Most cloud bills we inherit are larger than they need to be, and the first thing we do is find out exactly where the money goes.

What we hear before we start
Our cloud bill has tripled in two years and nobody can explain which service caused it.
We are still running production on a server under someone's desk. We know. Please help.
The migration has been six months from done for the last year.
What you get
  • Migration plan with rollback strategy
  • Infrastructure defined as code
  • Cost audit with itemised findings
  • Monitoring and alerting setup
  • Disaster recovery runbook
  • Documented landing zone
Typical tools
AWSAzureGCPTerraformDockerKubernetesCloudWatch
Discuss this practice
6–16 weeks
ComputeStorageNetwork
Typical duration
6–16 weeks
Team size
2–3 engineers
Commercial model
Fixed or time & materials
Downtime target
Zero for planned cutovers
04

Data & Analytics

Pipelines and dashboards your team actually opens. Most reporting projects fail because they answer questions nobody asked — so we start with the decisions you make every week, then build backwards to the data those decisions genuinely need.

What we hear before we start
Finance and operations produce different revenue numbers for the same month. Both are convinced they are right.
We built a dashboard last year. I do not think anyone has opened it since March.
Month-end reporting takes one person most of a week and it is always late.
What you get
  • Data pipeline with validation rules
  • Warehouse schema and documentation
  • Operational and executive dashboards
  • Automated scheduled reporting
  • Data quality monitoring
  • Audit trail on every figure
Typical tools
PythonAirflowSnowflakePostgreSQLdbtMetabase
Discuss this practice
6–14 weeks
IngestModelReport
Typical duration
6–14 weeks
Team size
2–3 engineers
Commercial model
Dedicated pod
Success measure
Weekly active dashboard users
05

Digital Transformation

Replacing manual and legacy processes in stages, with the old system running until the new one has earned its retirement. Big-bang cutovers are the single most common reason transformation programmes fail — we run both in parallel until the numbers agree.

What we hear before we start
Half our process still runs on paper forms that get typed in twice.
The last transformation project was abandoned after eighteen months and a lot of money.
Our staff have said outright they will not use another new system.
What you get
  • Current-state process map
  • Phased migration plan
  • Parallel-run reconciliation reports
  • Data migration with validation
  • Staff training materials
  • Adoption tracking after go-live
Typical tools
Process mappingChange managementData migrationIntegration design
Discuss this practice
12–36 weeks
MapMigrateAdopt
Typical duration
12–36 weeks
Team size
3–5 across phases
Commercial model
Phased programme
Cutover approach
Parallel run, never big-bang
06

Managed IT & Support

Monitoring, patching, and incident response for systems already in production — whether we built them or inherited them from someone else. A contracted SLA, a support line that actually picks up, and quarterly reviews where we tell you honestly what is ageing badly.

What we hear before we start
When it breaks at night, nobody knows who to call.
We inherited this system from a vendor who has stopped replying to emails.
Our dependencies have not been updated in three years and we are nervous about it.
What you get
  • 24/7 monitoring with escalation
  • Security patching schedule
  • Contracted response SLAs
  • Written incident reports
  • Quarterly service review
  • Prioritised improvement backlog
Typical tools
DatadogSentryPagerDutyTerraformGitHub Actions
Discuss this practice
Rolling annual
MonitorPatchRespond
Commitment
Rolling annual
Team size
Named on-call engineers
Commercial model
Retainer + SLA
Median P1 response
4 hours (SLA: 8)
Whichever practice you start with, the same team stays with you.
Engagement models

Three ways to work with us

Different problems need different commercial shapes. We recommend the one that fits, not the one that bills most.

Fixed-scope projectDedicated podManaged retainer
Best whenRequirements are clear and stablePriorities shift as you learnThe system is already live
BillingOne agreed priceMonthly per podMonthly retainer + SLA
Typical length8–24 weeks3 months minimumRolling annual
Scope changesVia written change orderFreely, each sprintWithin agreed capacity
Named teamYesYesYes, incl. on-call
Response SLABusiness hoursBusiness hours24/7 contracted
ExitOn delivery30 days notice60 days notice
Honest limits

What we don't do

Knowing where a firm stops is more useful than another list of things it claims to be good at.

Technology

What we build with

We pick tools for how long they will stay maintainable, not for how new they are.

ReactNext.jsTypeScriptNode.jsPythonDjangoPostgreSQLAWSAzureKubernetesTerraformReact NativeGoRedisKafkaSnowflake ReactNext.jsTypeScriptNode.jsPythonDjangoPostgreSQLAWSAzureKubernetesTerraformReact NativeGoRedisKafkaSnowflake
Questions

About working with us

Yes, and most clients do. A common path is a short consulting engagement, then development, then a managed retainer once the system is live. Each stage stands on its own — you are never committed to the next one.

Often. We can run as a self-contained pod alongside your team, or embed with them on a shared codebase. What we ask for is clarity on who owns which decisions, agreed before work starts.

On a fixed-scope project we raise a written change order so cost and timeline impact are visible before anything moves. On a dedicated pod, changing direction each sprint is the point of the model — no change orders needed.

This is the most important question for any shop-floor system, and the reason we pilot on one line first. Screens use large buttons, dropdowns instead of typing, and local language. If operators are still avoiding it after two weeks, that is a design problem on our side and we fix it before rolling further.

No. A shared Android tablet per line and printed badge cards are enough for a pilot. Sensors, machine data feeds, and biometric attendance are all possible later, but they are upgrades rather than requirements — and we would rather prove the system works before you spend on hardware.

You do, in full, from the first commit. You get repository access throughout the build rather than a handover at the end, along with infrastructure definitions and documentation.

Tell us the symptom, not the solution

You don't need a specification to start a conversation. Describe the problem and we'll tell you what kind of engagement actually fits.