Job Description

Key Responsibilities

• Develop, evaluate, optimize, and deploy machine learning models supporting real-time or operational incident management use cases, such as anomaly detection, predictive classification, forecasting, or computer vision.
• Manage the ML model lifecycle, including data preparation, model development, evaluation, deployment, monitoring, retraining, and version management.
• Design and develop Agentic AI systems using LLM orchestration frameworks such as LangChain, LangGraph, or equivalent, including multi-agent coordination, routing, state management, tool use, and multi-step reasoning.
• Develop Agentic AI solutions for real-time or operational incident management, including automated decision support and intelligent workflow automation.
• Integrate and fuse heterogeneous data sources, including structured and unstructured data, batch and streaming data, geospatial data, sensors, CCTV/video, telemetry, and crowdsourced feeds.
• Develop and apply appropriate data fusion methodologies to extract reliable signals and insights from multiple real-time data sources.
• Implement and maintain MLOps practices and infrastructure, including model deployment pipelines, automated retraining, model monitoring, drift detection, and deployment standards.
• Define and monitor measurable outcomes for ML, Agentic AI, and data fusion solutions, including model performance, latency, automation rate, data throughput, and operational improvements.

What we are Looking for

Experience

• At least 3 years of relevant experience as a Data Scientist, Machine Learning Engineer, AI Engineer, or equivalent.
• Experience delivering data science and AI solutions in production, proof-of-concept (POC), or proof-of-value (POV) environments.

Machine Learning & Model Development

• Hands-on experience developing, deploying, and optimising at least 2 machine learning models in a production, POC, or POV environment, with quantifiable performance outcomes using appropriate metrics (e.g. accuracy, precision, recall, F1-score, latency, or RMSE).
• Experience applying machine learning to real-time or operational incident management, such as anomaly detection, predictive classification, forecasting, or computer vision on live data streams.
• Experience with MLOps, including building automated retraining pipelines or model drift monitoring systems, with ownership of deployment standards and tooling decisions (e.g. MLflow, Kubeflow, SageMaker) and quantifiable outcomes evidenced (e.g. model uptime, retraining frequency, deployment turnaround time).

Agentic AI System Design & Development

• Hands-on experience designing and deploying at least 2 agentic AI systems incorporating LLM orchestration (e.g. LangChain, LangGraph, or equivalent frameworks) in a production, POC, or POV environment, with quantifiable outcomes evidenced (e.g. task automation rate, reduction in manual intervention, latency benchmarks met).
• Experience designing multi-agent architectures incorporating agent coordination, routing, state management, tool use, and multi-step reasoning, applied to a real-time or operational incident management use case.

Data Fusion & Heterogeneous Data Sources

• Hands-on experience integrating at least 2 heterogeneous data sources (e.g. structured/unstructured, batch/streaming) across at least 2 projects in a production, POC, or POV environment, with quantifiable outcomes evidenced (e.g. data throughput, data latency reduction, fusion accuracy, or volume of data sources integrated).
• Experience working with specialised data types such as geospatial data, sensor data, CCTV/video, or real-time telemetry streams, with clear description of the data fusion methodology applied.

Added Advantage

• Experience with operations, control centre, or incident-management systems.
• Experience deploying ML and Agentic AI solutions using cloud and containerised environments.
• Experience working with government or public-sector data and systems.