AI LLM Engineer - Autonomous Network

Capgemini; Abingdon; Hybrid; GBP 75,000 - 120,000; Full time

Benefits offered by this job

Hybrid working model

Job summary

Capgemini UK is hiring an AI/ML engineer to design and build the AI brain for autonomous network operations. You will develop LLM-based agents, RAG systems, multi-agent workflows, semantic search, and closed-loop decision support.

Bring strong Python skills and experience with LangChain, vector databases, graph databases, and cloud-native deployments. You will optimize latency, cost, and reliability while ensuring security and governance.

Qualifications

• Hands-on experience with LLMs, RAG, semantic search or agentic AI systems.
• Strong Python programming skills.
• Experience with ML fundamentals, deep learning concepts, embeddings, transformers and LLM architectures.
• Experience using LangChain, LangGraph, LlamaIndex, AutoGen, MCP or similar AI frameworks.
• Experience with vector databases such as Pinecone, Weaviate, Milvus, Qdrant, ChromaDB or equivalent.
• Experience with graph databases, knowledge graphs or Graph APIs.
• Experience building data pipelines and integrating structured and unstructured data sources.
• Understanding of AIOps, fault correlation, KPI modelling, predictive analytics or telecom network operations.
• Experience deploying AI services using Kubernetes, Docker, APIs and cloud‑native environments.

Responsibilities

• Design and develop LLM-based and agentic AI solutions for autonomous network operations.
• Build RAG frameworks using network documentation, alarms, topology, inventory, KPIs, trouble tickets, procedures, configuration data, and operational knowledge.
• Develop multi-agent workflows using LangChain, LangGraph, MCP, or similar frameworks.
• Implement vector search, semantic retrieval, graph-enhanced retrieval, and hybrid search patterns.
• Develop AI agents for fault diagnosis, root-cause analysis, KPI analysis, configuration recommendation, incident summarisation, and operational decision support.
• Build token-efficient prompting, context optimisation, caching, and response generation techniques.
• Integrate LLM solutions with OSS, AIOps, inventory, graph databases, vector databases, data pipelines, and automation platforms.
• Develop fault correlation, KPI modelling, predictive analytics, and closed-loop trigger logic.
• Implement safe AI workflows with human-in-the-loop approval, confidence scoring, explainability, and auditability.
• Optimise AI models and agent workflows for latency, cost, accuracy, and reliability.
• Support model evaluation, prompt evaluation, hallucination reduction, retrieval quality improvement, and grounding validation.
• Work with cybersecurity teams to implement LLM security, prompt injection protection, data leakage prevention and access controls.
• Deploy AI services using Kubernetes, Docker, APIs and cloud‑native patterns.

Skills

Python

ML basics

LLMs & transformers

LangChain & LangGraph

MCP or similar frameworks

Vector databases

Graph APIs

Data pipelines

Docker & Kubernetes

Fault correlation & KPI modelling

Predictive analytics & AIOps

Prompt engineering

OpenAI security & governance

Tools

LangChain

LangGraph

LlamaIndex

AutoGen

MCP

Pinecone

Weaviate

Milvus

Qdrant

ChromaDB

Kubernetes

Docker

APIs

Job description

Your Role

This role focuses on designing and building the AI brain for autonomous network operations. The AI / LLM Engineer will develop LLM-based agents, RAG systems, multi-agent workflows, semantic search, graph-enhanced reasoning, predictive analytics, KPI models, fault correlation models, and closed-loop decision support capabilities.

Key Responsibilities

• Design and develop LLM-based and agentic AI solutions for autonomous network operations.
• Build RAG frameworks using network documentation, alarms, topology, inventory, KPIs, trouble tickets, procedures, configuration data, and operational knowledge.
• Develop multi-agent workflows using LangChain, LangGraph, MCP, or similar frameworks.
• Implement vector search, semantic retrieval, graph-enhanced retrieval, and hybrid search patterns.
• Develop AI agents for fault diagnosis, root-cause analysis, KPI analysis, configuration recommendation, incident summarisation, and operational decision support.
• Build token-efficient prompting, context optimisation, caching, and response generation techniques.
• Integrate LLM solutions with OSS, AIOps, inventory, graph databases, vector databases, data pipelines, and automation platforms.
• Develop fault correlation, KPI modelling, predictive analytics, and closed-loop trigger logic.
• Implement safe AI workflows with human-in-the-loop approval, confidence scoring, explainability, and auditability.
• Optimise AI models and agent workflows for latency, cost, accuracy, and reliability.
• Support model evaluation, prompt evaluation, hallucination reduction, retrieval quality improvement, and grounding validation.
• Work with cybersecurity teams to implement LLM security, prompt injection protection, data leakage prevention and access controls.
• Deploy AI services using Kubernetes, Docker, APIs and cloud-native patterns.

Your Profile

• Experience in AI/ML engineering, data engineering, software engineering or applied machine learning.
• Hands‑on experience with LLMs, RAG, semantic search or agentic AI systems.
• Strong Python programming skills.
• Experience with ML fundamentals, deep learning concepts, embeddings, transformers and LLM architectures.
• Experience using LangChain, LangGraph, LlamaIndex, AutoGen, MCP or similar AI frameworks.
• Experience with vector databases such as Pinecone, Weaviate, Milvus, Qdrant, ChromaDB or equivalent.
• Experience with graph databases, knowledge graphs or Graph APIs.
• Experience building data pipelines and integrating structured and unstructured data sources.
• Understanding of AIOps, fault correlation, KPI modelling, predictive analytics or telecom network operations.
• Experience deploying AI services using Kubernetes, Docker, APIs and cloud‑native environments.

Required Technical Skills

• Python
• ML basics and deep learning
• LLMs and transformers
• LangChain, LangGraph, MCP or similar frameworks
• Vector databases and semantic search
• Graph APIs and knowledge graphs
• Data pipelines and data aggregation
• Docker and Kubernetes
• Fault correlation and KPI modelling
• Predictive analytics and AIOps
• Closed‑loop triggers
• Prompt engineering and context optimisation
• AI observability and evaluation

Preferred Certifications

• Google Cloud AI/ML or Vertex AI certification
• Azure AI Engineer or AWS Machine Learning certification
• Databricks, BigQuery or data engineering certification
• Kubernetes certification
• TM Forum Autonomous Networks or Open API certification

Nice‑to‑Have Qualifications

• Experience with Google Vertex AI, Gemini APIs, BigQuery or equivalent platforms
• Experience with telecom network data including RAN, Core, IP/MPLS, SD‑WAN, OSS, alarms, KPIs and inventory
• Experience developing LLM agents for network operations, incident management or service assurance
• Experience with AI model optimisation, inference cost reduction, latency optimisation and scalable AI serving

If you’re excited about this role but don’t meet every requirement, we still encourage you to apply; your unique experience could be just what we need.

Make it real – what does it mean for you?

• Exposure to top global companies working with Capgemini (145 of the Fortune 500 companies)
• Open access to digital learning platforms
• Active employee networks promoting diversity, equity and inclusion like OutFront, CapAbility or Women@Capgemini

Eligibility

Capgemini is proud to be a Disability Confident Employer (Level 2) under the UK Government’s Disability Confident scheme. As part of our commitment to inclusive recruitment, we will offer an interview to all candidates who:

• Declare they have a disability, and
• Meet the minimum essential criteria for the role.
• Please opt in during the application process.

Need to know

• All roles will require a level of security clearance; BPSS OR Security Clearance OR Developed Vetting.
• Location: This is a permanent role with Capgemini, offering a hybrid working model. The client is based in Newbury and occasional travel to the client site will be required.
• You can bring your whole self to work. At Capgemini building an inclusive future is part of everyday life and will be part of your working reality. We have built a representative and welcoming environment, for everyone.