Autonomous Network AI Scientist (Hybrid)

Capgemini Engineering

Greater London

Hybrid

GBP 100,000 - 130,000

Full time

14 days+

Application generator

Turn this role into an interview — a resume and cover letter built around what this employer wants.

Get past ATS filters

Benefits offered by this job

Exposure to top global telco projects

Digital learning platforms access

Diversity and inclusion networks

Job summary

Capgemini is seeking an AI Data Scientist to apply data science, statistics, ML, graph analytics and KPI engineering to enable autonomous network intelligence across telecom domains. You will work with network telemetry, alarms, inventory, topology, and digital twin models to deliver analytical insights and predictive intelligence for autonomous operations.

You will develop KPI models, anomaly detection and proactive assurance, build data pipelines, and collaborate with AI/LLM engineers to

Qualifications

Experience in data science, telecom analytics, network performance analytics, or AIOps analytics.
Strong Python skills for data analysis, modelling, and automation.
Strong knowledge of statistics, ML basics, feature engineering, and model evaluation.
Experience working with network KPIs, alarms, telemetry, inventory, topology, or service assurance data.
Understanding of telecom network domains including RAN, Core, IP/MPLS, SD‑WAN, Transport, Cloud, or OSS.
Experience developing anomaly detection, fault analysis, predictive analytics, or KPI models.
Experience with data aggregation, cleansing, enrichment, and data quality assessment.
Experience with graph analytics, entity relationship mapping, or knowledge graph concepts.
Understanding of LLMs and how structured data can support AI agents and RAG systems.
Experience with BigQuery or similar data warehouse/analytics platforms.

Responsibilities

Analyse large‑scale telecom network datasets across RAN, Core, IP, Transport, SD‑WAN, Cloud, OSS, and service domains.
Develop KPI engineering models for network performance, service quality, fault behaviour, customer impact, capacity, and resilience.
Build statistical and machine learning models for anomaly detection, fault prediction, root‑cause analysis, degradation detection, and proactive assurance.
Develop data aggregation, cleansing, enrichment, and feature engineering pipelines for network telemetry and OSS data.
Support digital twin analytics using topology, inventory, configuration, service dependency, performance, and fault data.
Develop graph analytics models for network topology, entity relationships, dependency mapping, service impact, and fault propagation.
Work with AI/LLM engineers to provide high‑quality features, embeddings, metadata, and contextual datasets for RAG and agentic AI systems.
Define network data quality rules, correlation logic, and entity resolution methods.
Create reusable analytical models for RAN, Core, IP/MPLS, SD‑WAN, fixed, and cloud network KPIs.
Support AIOps use cases such as alarm reduction, incident prioritisation, predictive maintenance, and automated root‑cause analysis.
Work with OSS and inventory teams to align data models with TMF SID concepts and TMF Open API structures.
Use BigQuery or equivalent analytics platforms to process large‑scale network data.
Ensure models are explainable, measurable, governed, and suitable for operational decision‑making.