Skip to content
nxtinno

Case study

IoT and energy data platform

Designed data ingestion and analysis workflows for distributed meters and sensors, combining device integration, time-series data, automation, dashboards, anomaly detection, and AI-assisted insights.

  • MQTT
  • Time-series DB
  • Node.js
  • Automation

Context

An energy-sector environment with distributed meters and sensors producing continuous measurements that needed to become dashboards, alerts, and decisions rather than raw data.

Challenge

Device data arrived through heterogeneous protocols with gaps, duplicates, and late readings. Turning it into trustworthy time series — and then into automated insight — required an architecture, not another dashboard tool.

Constraints

  • heterogeneous devices and protocols in the field
  • unreliable connectivity with late and duplicate readings
  • time-series volumes that grow continuously and never stop
  • operational users who need alerts they can trust

Approach

  • Designed a device-integration layer that normalised protocols and tolerated disconnects without losing readings.
  • Modelled measurements as time series with explicit quality flags so downstream analytics knew what they were standing on.
  • Built automation workflows for threshold and anomaly-based alerting, with humans confirming actions where consequences were physical.
  • Layered AI-assisted analysis on top of validated data — insight generation only after the deterministic pipeline was dependable.

Architecture

MQTT-based device ingestion into a time-series store, stream processing for validation and alerting, workflow automation for operations, and an analysis layer providing AI-assisted insights over validated data.

Production RAG and agent workflowSource systemsdocs · data · APIsIngestionnormalise · chunkRetrievalhybrid · filteredModel + toolsMCP · scoped accessGuardrailseval · approvalDependable outputobserved · attributed
Diagram summary: source systems feed an ingestion and chunking pipeline; retrieval and context assembly ground the model, which acts through scoped MCP tools; evaluation and guardrails observe every step before an answer or action reaches the user.

Key decisions

  • data quality made explicit in the model instead of assumed
  • deterministic pipeline first, AI-assisted insight second
  • automation with human confirmation for physically consequential actions

Details are anonymised. Additional context is available under NDA.

Related expertise

AI systems and agentic workflowsProduct engineering

Curious what AI could actually do for your business?

Bring your questions, including the ones that feel too basic. In 30 minutes we go through how you work today, pick the task with the most to gain, and sketch what testing it would involve. No pitch, no obligation.

Book a free intro call

Prefer email? hello@nxtinno.com