FULL-TIME · ONSITE · HYDERABAD
Product Engineering / AI & Digital Solutions
About Refroid
Refroid is building next-generation sustainable cooling technologies for AI data centers, industrial cooling, and high-performance computing environments. Our solutions combine advanced cooling hardware with intelligent software to optimize energy consumption, maximize equipment life, and proactively manage mission-critical infrastructure. As Refroid expands globally, we are building an AI-powered digital intelligence layer that turns operational data into predictive insights and autonomous recommendations.
Role Purpose
The role requires strong embedded-system expertise to interface with PLCs, microcontrollers, industrial sensors, and edge devices, enabling reliable data acquisition, protocol integration, edge intelligence, and seamless flow of operational data into cloud analytics platforms.
Key Responsibilities
Build the Data Intelligence Platfrom
Design end-to-end data pipelines connecting IoT sensors, PLCs, controllers, edge devices, and cloud platforms.
Develop robust ingestion pipelines for real-time streaming data.
Ensure data quality, cleansing, validation, synchronization, and standardization.
Manage historical operational datasets for long-term analytics.
Develop software modules to acquire, decode, validate and normalize data from PLCs, Microcontroller, embedded controllers, Industrial sensors and custom electronic hardware. Implement industrial communication protocols (Modbus, TCP/IP, RS-485, SNMP, SPI, I2C, I3C, etc.) for reliable data acquisition.
Develop robust ingestion pipelines for real-time streaming data (control plane and data plane).
Work closely with firmware and embedded teams to define sensor telemetry. sampling strategies, edge processing and communication interfaces.
Support embedded software architecture for reliable sensor integration and data acquisition across Refroid products.
Machine Learning & Predictive Analytics
Develop ML models capable of:
Predictive maintenance
Early failure detection
Equipment health scoring
Cooling performance optimization
Thermal anomaly detection
Pump and fan performance prediction
Heat exchanger efficiency prediction
Energy consumption forecasting
Cooling capacity prediction
Remaining Useful Life (RUL) estimation
Sensor drift detection
Operational risk prediction
Product Intelligence
Convert millions of sensor readings into business intelligence by identifying:
System inefficiencies
Performance degradation
Emerging failure patterns
Operational bottlenecks
Abnormal equipment behavior
Site-level performance trends
Fleet-wide benchmarking
Customer usage patterns
AI-Driven Decision Support
Design intelligent recommendation engines that can answer questions such as:
Design intelligent recommendation engines that can answer questions such as:
Which cooling unit is likely to fail in the next 30 days?
Which site consumes more energy than expected?
Which pump requires servicing?
Which component is degrading?
Why has thermal efficiency dropped?
What operational changes can improve performance?
What preventive actions should be scheduled?
Dashboard & Visualization Develop intelligent dashboards that provide:
Executive Dashboard
Fleet health score
Cooling efficiency
Energy savings
Carbon reduction
System availability
Installed base overview
Engineering Dashboard
Thermal performance
Pressure distribution
Flow analytics
Sensor health
Equipment diagnostics
Root cause analysis
Customer Dashboard
Live monitoring
Asset health
Predictive maintenance alerts
Recommended actions
Service notifications
Energy optimization insights
Product Enhancement
Collaborate with Product Engineering to:
Improve sensor architecture
Recommend additional instrumentation
Enhance data acquisition methods
Improve firmware data collection
Define next-generation AI-enabled cooling products
Drive continuous product improvement using operational insights
Cross Functional Collaboration Work closely with:
Thermal Engineering
Embedded Systems
Electronics
Firmware
Product Design
Manufacturing
Customer Success
Service Engineering
Digital Products
Required Technical Skills
Data Engineering: Python, SQL, time-series databases, ETL pipelines, streaming data, REST APIs, MQTT, OPC-UA, and edge computing.
Machine Learning: scikit-learn, TensorFlow, PyTorch, XGBoost, LSTM, transformers, autoencoders, anomaly detection, regression, classification, forecasting.
Data Analytics: statistical analysis, signal processing, feature engineering, root-cause analysis, and predictive modelling.
Visualization: Power BI, Grafana, Tableau, Plotly, and Streamlit.
Cloud Platforms: Azure, AWS, and Google Cloud.
IoT: industrial sensors, embedded devices, PLCs, SCADA, IoT gateways, and Edge AI.
Preferred Domain Experience
Experience in one or more of Industrial IoT, HVAC, cooling systems, thermal engineering, semiconductor manufacturing, data centers, smart manufacturing, energy analytics, renewable energy, or industrial automation.
Qualifications
Master’s or Ph.D. in Computer Science, Data Science, Machine Learning, Electronics, Electrical Engineering, Instrumentation, Mechanical Engineering, or related disciplines.
5–10 years of experience in Machine Learning, Data Engineering, or AI product development.
Proven experience handling large-scale sensor or time-series data.
Experience deploying production-grade ML models.
Strong understanding of statistical modeling and predictive analytics.
Success Metrics — First 12 Months
Build Refroid’s unified sensor data platform.
Develop predictive models with measurable accuracy for anomaly detection and failure prediction.
Deliver executive, engineering, and customer dashboards.
Reduce unplanned service incidents through predictive maintenance.
Improve cooling system efficiency using AI-driven recommendations.
Enable data-informed product improvements based on field performance.
Establish a scalable analytics foundation for all deployed Refroid cooling systems.
Closing Remark
This is not a conventional Machine Learning position. It combines embedded system engineering, industrial IoT, data engineering, AI, and product analytics to build an end-to-end intelligent cooling platform from sensor acquisition through autonomous decision-making.