WhatsApp
Tecifo Solutions logo: orange hexagon cube with blue hexagon dotsTecifo.Book a call
← All work
IoTprivate client

Energy Monitoring System

Cloud and on-premise telemetry platform ingesting live device data with real-time consumption dashboards.

Sector
Energy / industrial
Engagement
Build — monitoring platform
Timeline
Aug 2023 → Mar 2024
Team
1 backend engineer, 1 full-stack engineer
01 — The problem

An industrial operator was paying for energy it could not attribute to anything.

Meters existed across the operator's facilities, but readings were collected monthly by a technician with a clipboard and aggregated into one figure per site. Nobody could say which line, floor or piece of equipment was responsible for a spike, so efficiency projects were argued about rather than measured.

Some facilities could not send data off-premise for policy reasons, which ruled out a cloud-only product. The same platform had to run in the operator's own environment for those sites and in the cloud for the rest, without maintaining two codebases.

02 — How we solved it

What we built, and why in that order

01
MQTT ingestion
Meters publish to a broker and a Django ingestion service normalises and stores readings, handling the intermittent connectivity and duplicate publishes that industrial devices produce.
02
Consumption dashboards
Live and historical consumption by site, line and device, with baselines so an anomaly is visible as a departure from normal rather than a raw number.
03
One build, two deployments
Docker Compose for on-premise, AWS for the cloud tenants, from the same image and migrations. Air-gapped sites get the same features as connected ones.
04
Reporting on MySQL
A schema designed for time-series queries at the granularity finance and engineering each needed, so monthly reporting reads from the same data the dashboards do.
StackDjangoPythonMySQLMQTTAWSDocker
Outcome
monthly → 5s
Reading interval
Device-level
Attribution, previously site-level
2
Deployment targets, one codebase
9%
Consumption reduction in first two quarters
For the first time an efficiency claim comes with a chart instead of an argument.
Plant engineering lead, energy client
03 — Where they are now

The operator now measures efficiency work instead of estimating it.

Our role: Telemetry platform engineering

Baseline comparisons made two long-running assumptions falsifiable in the first quarter: one machine was drawing far more at idle than anyone believed, and a scheduled process was running outside its window. Both were fixed from dashboard evidence, and the consumption reduction that followed is attributed to specific equipment.

The platform has expanded to further facilities and now feeds the operator's sustainability reporting directly. Their internal team runs the deployments; we handle new device integrations and schema changes as the estate grows.

Next case studyStarling MXAI/ProductSimilar problem?Start a projectinfo@tecifo.com

Have something that needs to ship?

Scope agreed in writing, architecture first, weekly increments you can review. Start with a 30-minute call or send the brief.