About the Opportunity:
Ericsson Enterprise Wireless Solutions (BEWS) is responsible for driving Ericsson's Enterprise Networking and Security business. Our expanding product portfolio covers wide area networks, local area networks, and enterprise security. We are the #1 global market leader in Wireless-WAN enterprise connectivity and are rapidly growing in enterprise Private 5G networks and Secure Access Services Edge (SASE) solutions.
What Will You Do:
Lead the design and implementation of machine learning and statistical models to solve complex business problems across multiple domains.
Architect and build Agentic AI systems - including multi-agent pipelines, autonomous reasoning loops, tool-using agents, and agent orchestration to enable proactive, self-directed network operations.
Design and deploy LLM-powered applications including RAG pipelines, fine-tuned domain models, prompt engineering strategies, structured output generation, and evaluation frameworks (LLM-as-judge, RAGAS, etc.).
Collaborate with cross-functional teams - data engineers, product managers, and architects - to identify high-impact opportunities for AI-driven solutions and translate them into production-grade systems.
Perform data exploration, feature engineering, and dataset curation to support both classical ML model development and LLM grounding/context pipelines.
Evaluate model performance using rigorous metrics, and ensure interpretability, fairness, and robustness across model types - from tree-based models to transformer architectures.
Contribute to reusable AI frameworks, internal tooling, and shared libraries to raise engineering velocity across the team.
Stay at the forefront of AI/ML research - actively tracking arXiv, open-source releases, and industry developments
What You Will Bring:
8–14 years of applied Data Science and Machine Learning experience, with demonstrated progression in scope and complexity of owned systems.
Deep expertise in classical ML - regression, classification, clustering, ensemble methods, time-series modeling, and anomaly detection - with strong intuition for when to apply them.
Hands-on experience building LLM-based systems: RAG architectures, prompt engineering, context window management, vector databases, and LLM orchestration frameworks
Practical experience designing Agentic AI workflows - including tool-calling agents, memory architectures
Proficiency in model fine-tuning techniques: LoRA, QLoRA, PEFT, instruction tuning, and RLHF/DPO alignment approaches.
Strong programming skills in Python, with proficiency in SQL and familiarity with distributed data processing
Solid software engineering discipline - clean, well-tested, reproducible code with CI/CD awareness.
Experience deploying and scaling AI models in production with robust monitoring, observability, and performance management.
Proven track record of end-to-end ownership of AI/ML projects - from problem framing through production deployment and iteration.
Preferred Qualifications:
Master's or PhD in Computer Science, Statistics, Mathematics, or a related field.
Experience with deep learning frameworks (PyTorch, TensorFlow) and training or fine-tuning large models.
Familiarity with model safety, guardrails, and responsible AI practices, including output filtering, hallucination mitigation, and bias evaluation.
Experience with function calling / tool use patterns in LLMs and building structured, reliable agent behaviors.
Domain exposure in telecom, network operations, or AIOps is highly desirable.
Contributions to open-source AI projects or published research is a plus.
“All academic credentials must be from recognized and accredited institutions and are further subject to verification.”
