Arbetsbeskrivning

Ericsson AB

About this opportunity:

The AI Accelerator within BNEW SAN is on a mission to innovate, incubate, and deliver AI-based solutions that fundamentally redesign Customer Support and Network Deployment — shifting from human-led coordination to composable, agent-centric intelligence. We aim to move beyond simple efficiency gains to create a compounding cycle of optimization, where every interaction generates actionable insights that continuously refine operations, reduce time-to-market, and accelerate the journey toward autonomous network operations.

We are looking for a Lead AI Systems Architect to own the end-to-end technical vision and system coherence of our Autonomous Multi-Agent System. This role bridges architecture, AI strategy, and engineering execution — setting the blueprint that all engineering roles build toward and driving the long-term evolution of the system through model advances, domain expansion, and operational scale.

You will be the definitive technical authority across agent topology, orchestration design, evaluation frameworks, and capability governance. You will work closely with engineering leads, domain partners, product management, and executive stakeholders to ensure that every architectural decision serves both near-term delivery and long-term strategic ambition.

What you will do:

Architecture Design

• Define the canonical agent topology: roles, boundaries, interaction contracts, and trust model
• Own the reference architecture for orchestration, memory, tool calling, and inter-agent communication
• Establish architectural decision records and enforce consistency across all subsystems
• Design the observability and governance layer: tracing, audit trails, and agent accountability

System Design

• Translate architecture into concrete system specifications: data flows, API contracts, failure modes, and SLA bounds
• Define agent skill taxonomy, system prompt standards, and memory management policy
• Own the evaluation and benchmarking framework design across all agent roles
• Ensure domain-specific requirements — including RAN, Core Network, and broader customer support and network deployment telco constraints — are encoded into system design

Strategy & Planning

• Define the product vision and technology roadmap, balancing research ambition with deployment reality
• Track and evaluate emerging LLM and agentic AI research; translate relevant advances into concrete adoption decisions
• Align technical direction with business objectives, domain partner requirements, and platform constraints
• Lead build vs. adopt vs. integrate decisions across base models, tooling, and infrastructure

System Evolution

• Own the agentic system capability lifecycle: version governance, deprecation policy, and backward compatibility
• Drive continuous improvement loops: evaluation → post-training → deployment → re-evaluation
• Manage base LLM upgrade cycles and assess downstream impact across all agent roles
• Identify and retire technical debt in agent design, memory architecture, and tool interfaces