How Transparent Path Built a Real-Time IoT Tracking Platform for Perishables Supply Chains with Kubernetes and Apache Kafka

Client

Transparent Path is a supply chain intelligence company focused on the perishables sector — the segment of logistics where product quality is time-critical and the cost of information gaps is measured in wasted inventory, food safety incidents, and lost producer revenue. Transparent Path’s platform premise is that real-time sensor data and automated alerting can transform how perishables networks respond to quality events: shifting from reactive, phone-based problem resolution to proactive, data-driven intervention. The engagement with Ajackus was conducted through a Buildly open-source collaboration, with Ajackus taking engineering ownership of the IoT integration, cloud architecture, and data pipeline components of the platform.

Challenge

The perishables supply chain is one of the most information-intensive logistics environments in existence. Produce, dairy, pharmaceutical products, and other temperature-sensitive goods move through six distinct stakeholder tiers — producers, processors, logistics providers, wholesalers, distribution hubs, and retailers — and the quality window for intervention is often measured in hours, not days. When a temperature excursion occurs in transit and the first notification comes via a phone call from a driver or warehouse worker, the damage is often already done.

Transparent Path identified this gap as both a commercial opportunity and an engineering challenge: the sensor hardware to monitor shipment conditions existed, but no platform had integrated it into a complete supply chain intelligence system that could act on sensor data at the speed the perishables industry requires.

Manual, Paper-Based Problem Resolution

Incident response across the perishables supply chain relied on phone calls, spreadsheets, and manual data entry. When a quality event occurred — a temperature excursion, a humidity spike, a transit delay — the information propagated through the network at human speed: one phone call at a time, with no structured record, no automated escalation, and no way to coordinate consortium-wide response. The result was slow intervention, inconsistent documentation, and preventable losses.

Significant Goods Wastage from Information Delays

Without real-time visibility into shipment conditions, subpar products regularly progressed further through the supply chain than they should — reaching wholesalers or retailers before the quality issue was identified. By that point, the remediation options were limited and the economic loss was already incurred. The absence of automated alerting meant that every quality event was discovered late, and late discovery is the primary driver of perishables waste.

Fragmented Data Across the Supply Network

Each tier of the supply chain maintained its own records in its own format, with no shared data layer to enable coordinated decision-making. Consortium-level risk management — the ability for producers, logistics partners, and wholesalers to collaborate in real time on a quality event — was operationally impossible without a common platform. The data gaps between tiers meant that the full picture of any incident only emerged after the fact, through laborious manual reconciliation.

Rising Fraud Risk and Regulatory Complexity

Global population growth and increasing regulatory scrutiny of food safety have raised the stakes for perishables traceability. The combination of inadequate real-time tracking, manual record-keeping, and fragmented network data created both a fraud exposure and a compliance risk — as regulators increasingly demand evidence of chain-of-custody and condition monitoring throughout the supply journey.

Goals

The platform needed to function as a complete supply chain intelligence layer — ingesting IoT sensor data, processing it in real time, alerting the right stakeholders automatically, and enabling consortium-wide coordination across all six tiers of the perishables network.

Journey

The Ajackus team took engineering ownership of the Transparent Path platform across four parallel workstreams — IoT sensor integration, real-time data pipeline, cloud infrastructure, and platform application development — working within the Buildly open-source collaboration framework. From the outset, the Ajackus team prioritised the data pipeline and sensor integration layers as the architectural foundation, recognising that the application features (alerting, consortium management, reporting) could only be as reliable as the data they were built on.

IoT Sensor Integration with ICLP and Tive

The Ajackus team integrated two IoT sensor platforms — ICLP and Tive — into the Transparent Path system, enabling real-time ingestion of temperature and humidity readings from devices deployed across the perishables supply chain. Each sensor platform has distinct data schemas, communication protocols, and event formats; the Ajackus team built a normalisation layer that translates readings from both platforms into a unified data model, ensuring that downstream alerting and analytics logic operates consistently regardless of which sensor hardware generated the data. This two-platform integration also gave the network operational flexibility — operators can deploy whichever sensor platform is appropriate for a given shipment type without any change to the platform logic.

Real-Time Data Pipeline with Apache Kafka

The volume and frequency of IoT sensor events across a multi-tier supply chain requires a data pipeline engineered for high-throughput, low-latency processing. The Ajackus team implemented Apache Kafka as the event streaming backbone, selected for its ability to handle large volumes of concurrent sensor streams with guaranteed message delivery and minimal processing lag. Kafka’s topic-based architecture also enables the platform to route events to multiple downstream consumers simultaneously — the alerting engine, the reporting system, and the monitoring dashboard — without any one consumer affecting another’s processing performance.

Kubernetes-Orchestrated Cloud Architecture on AWS and GCP

The Ajackus team designed the platform as a microservices system orchestrated by Kubernetes, with each functional domain — sensor ingestion, event processing, alert management, consortium coordination, report generation — deployed as an independently scalable service. Terraform and Helm were implemented for Infrastructure as Code, ensuring the platform can be replicated across environments through a version-controlled, parameterised process consistent with Buildly’s open-source distribution model. The multi-cloud architecture supporting both AWS and GCP was a deliberate resilience decision — the platform is not dependent on a single cloud provider’s availability or commercial terms.

Monitoring Infrastructure with Prometheus and Alerting Engine

The Ajackus team implemented Prometheus for real-time system monitoring across the platform’s microservices, providing operational visibility into system health, data pipeline throughput, and alert delivery status. Alongside infrastructure monitoring, the Ajackus team built the customisable alerting system — allowing supply chain operators to define deviation thresholds per shipment, per product type, or per route, with automated notifications delivered to the appropriate stakeholders when those thresholds are breached. The alerting engine distinguishes between minor deviations requiring adjustment and critical safety events requiring consortium-wide escalation, routing each to the appropriate response workflow.

Results

Transparent Path operates a production-ready supply chain intelligence platform that replaces manual, phone-based incident response with automated real-time alerting, consortium coordination, and structured reporting across the full perishables supply network.

What went well:

Operational Improvements

• Manual, phone-based incident resolution replaced by automated deviation alerts delivered to the appropriate stakeholders the moment sensor readings breach defined thresholds
• Consortium management capability enables producers, logistics partners, wholesalers, and retailers to coordinate on quality events through a shared platform interface rather than fragmented bilateral communication
• Automated report generation eliminates manual data entry from supply chain event documentation, creating a structured, searchable record of every tracked shipment
• Customisable alert thresholds allow operators to define deviation sensitivity per product type, shipment route, or client requirement — without any platform code changes

Technical Achievements

• ICLP and Tive IoT sensor platforms integrated through a normalised data layer, enabling consistent real-time temperature and humidity ingestion regardless of sensor hardware
• Apache Kafka event streaming pipeline handles high-volume concurrent sensor streams with guaranteed delivery and minimal processing lag across all connected supply chain tiers
• Kubernetes microservices architecture on AWS and GCP provides independent scalability for each platform function — sensor ingestion, alerting, consortium management, and reporting
• Prometheus monitoring delivers real-time visibility into system health, data pipeline throughput, and alert delivery performance across all microservices
• Terraform and Helm Infrastructure as Code enables consistent, version-controlled environment replication for Buildly’s open-source distribution model

Business Impact

• Subpar products are now intercepted earlier in the supply chain — automated alerts enable faster intervention before goods progress to stages where remediation is no longer commercially viable
• Critical safety events trigger consortium-wide alerts, ensuring that all network partners can act simultaneously rather than waiting for information to propagate through phone-based channels
• The platform’s cloud-portable, open-source architecture positions Transparent Path to expand network coverage and onboard new stakeholder tiers without infrastructure rearchitecting