Investigate agentic AI for secure and reliable telecom operations by building a controlled evaluation environment. Compare deterministic, language-model, tool-using, multi-agent, and hybrid architectures across network operations tasks. Evaluate reliability, security, safety, explainability, latency, cost, scalability, observability, and recovery. Design controls for prompt injection, tool misuse, privilege escalation, data leakage, and unsafe plans, then translate findings into architecture and adoption recommendations.
Join our Team
About this opportunity
Telecom networks are becoming more intelligent, automated, and software-defined. Agentic AI could help engineers and operations teams understand network state, investigate incidents, plan actions, and coordinate tools across complex domains. Before these capabilities can be trusted, we need to understand how to design agents that are correct, secure, explainable, observable, and scalable.
Ericsson Innovation Center EMEA is inviting one or two master thesis students to investigate agentic AI for telecom operations. You will compare different agent patterns, including deterministic workflows, probabilistic language-model agents, tool-using agents, multi-agent approaches, and hybrid solutions that combine AI with explicit mathematical models or optimization.
The project is intended to produce both an academic contribution and a tangible evaluation environment that can guide future innovation work.
What you will do
You will select one representative telecom operations problem and build a controlled environment in which alternative agent designs can be evaluated. You may work with a simulator, digital twin, synthetic network, or sandboxed lab environment.
Your work may include, but is not limited to:
The skills you bring
About this opportunity
Telecom networks are becoming more intelligent, automated, and software-defined. Agentic AI could help engineers and operations teams understand network state, investigate incidents, plan actions, and coordinate tools across complex domains. Before these capabilities can be trusted, we need to understand how to design agents that are correct, secure, explainable, observable, and scalable.
Ericsson Innovation Center EMEA is inviting one or two master thesis students to investigate agentic AI for telecom operations. You will compare different agent patterns, including deterministic workflows, probabilistic language-model agents, tool-using agents, multi-agent approaches, and hybrid solutions that combine AI with explicit mathematical models or optimization.
The project is intended to produce both an academic contribution and a tangible evaluation environment that can guide future innovation work.
What you will do
You will select one representative telecom operations problem and build a controlled environment in which alternative agent designs can be evaluated. You may work with a simulator, digital twin, synthetic network, or sandboxed lab environment.
Your work may include, but is not limited to:
- Defining a benchmark for troubleshooting, configuration recommendation, incident triage, capacity optimization, or another network task.
- Comparing deterministic automation, language-model agents, multi-agent designs, and hybrid architectures.
- Investigating how mathematical models, constraints, planning, or optimization can improve agent reliability.
- Designing security controls for prompt injection, tool misuse, privilege escalation, data leakage, unsafe plans, and cascading failures.
- Evaluating task success, correctness, constraint violations, safety, latency, cost, explainability, reproducibility, and recovery behavior.
- Studying human-in-the-loop approval, policy enforcement, auditability, rollback, and observability.
- Investigating what is required to scale an agent from an innovation prototype to a production-like service.
- Categorizing failures and translating the results into practical architecture and adoption recommendations
The skills you bring
- Knowledge in Computer Science, Computer Engineering, Artificial Intelligence, Data Science, Software Engineering, or a related field.
- Programming experience in Python, Java, Scala, TypeScript, or a similar language.
- Interest in generative AI, agentic systems, optimization, cybersecurity, distributed systems, or telecom networks.
- Experience with APIs, containers, evaluation frameworks, or cloud-native development is beneficial.
- Interest in designing experiments and analyzing quantitative results.
- Eagerness to learn, challenge assumptions, and communicate findings to both technical and non-technical stakeholders.
Similar Jobs at Ericsson
Cloud • Information Technology • Internet of Things • Machine Learning • Software • Cybersecurity • Infrastructure as a Service (IaaS)
Analyze large-scale Ericsson test and verification data, develop data-driven quality assessment and anomaly detection approaches, investigate AI and machine learning techniques, and prototype scalable cloud-native data platform solutions. The role includes evaluating release confidence models, supporting engineering decisions, and presenting findings.
Top Skills:
Analytics PlatformsAnomaly DetectionArtificial IntelligenceCloud-Native ArchitectureData PipelinesMachine LearningPython
Cloud • Information Technology • Internet of Things • Machine Learning • Software • Cybersecurity • Infrastructure as a Service (IaaS)
Develop and maintain zero-touch network provisioning and secure deployment automation across Ericsson test labs. Build integrations, validation processes, and end-to-end test capabilities; troubleshoot issues and improve reliability. Collaborate with engineers, architects, and stakeholders to deliver maintainable solutions. Use AI-assisted development and LLM tools for coding, testing, analysis, documentation, and knowledge sharing while contributing to continuous improvement.
Top Skills:
Ai-Assisted DevelopmentAutomated TestingC++Cloud-Native DevelopmentContainersContinuous IntegrationGitGitlabGitlab RunnerGoGroovyJenkinsLarge Language ModelsLinuxNautobotNetwork AutomationNetwork-Management SystemsPythonTcp/IpZero-Touch Provisioning
Cloud • Information Technology • Internet of Things • Machine Learning • Software • Cybersecurity • Infrastructure as a Service (IaaS)
Design, develop, integrate, and demonstrate technology testbed systems spanning algorithms, software architecture, distributed systems, real-time platforms, hardware-near software, and scalable software. Lead technical studies, invent new concepts and algorithms, collaborate across disciplines, and influence technology roadmaps through prototype development and evaluation. The role focuses on emerging mobile communication technologies, including radio design, signal processing, AI/ML, software, and compute platforms.
Top Skills:
Ai/MlAsynchronous SystemsDistributed SystemsHardware-Near SoftwareReal-Time SystemsScalable SoftwareSignal ProcessingSoftware ArchitectureTelecommunications
What you need to know about the London Tech Scene
London isn't just a hub for established businesses; it's also a nursery for innovation. Boasting one of the most recognized fintech ecosystems in Europe, attracting billions in investments each year, London's success has made it a go-to destination for startups looking to make their mark. Top U.K. companies like Hoptin, Moneybox and Marshmallow have already made the city their base — yet fintech is just the beginning. From healthtech to renewable energy to cybersecurity and beyond, the city's startups are breaking new ground across a range of industries.
