Learning to optimize. Building dependable systems.
I am an AI Research Engineer at Aerobase, working on neural combinatorial optimization and reinforcement learning.
During my PhD at Luleå University of Technology (2020–2025), I studied formal verification and AI agents for IEC 61499 control software. That work shaped my interest in systems that can reason, act, and be checked—a foundation I now bring to learning-based optimization at Aerobase.
Research Publications CV Contact
Research themes
Learning-based combinatorial optimization
I study attention-based neural solvers for routing, scheduling, and planning, including learned improvement heuristics and generalization across problem sizes and distributions.
Reinforcement learning and policy optimization
My interests include training neural solvers with reinforcement learning and best-of-K objectives that align training with how candidate solutions are selected at inference time.
Dependable automation and AI agents
My doctoral research connects formal methods with industrial software development. My publications explore verification, reasoning, planning, and testing for automation and agent workflows.
Selected work
ReACT-Gen AI Agents for Reasoning, Planning, and Testing in IEC 61499-Based Control Systems
IECON 2025 – 51st Annual Conference of the IEEE Industrial Electronics Society
This paper presents ReACT-Gen AI agents that combine reasoning, planning, and automated testing to support the design and validation of IEC 61499-based control systems.
IndustriConnect: MCP Adapters and Mock-First Evaluation for AI-Assisted Industrial Operations
arXiv preprint
This work introduces IndustriConnect, a set of Model Context Protocol (MCP) adapters together with a mock-first evaluation methodology for developing and assessing AI-assisted industrial operations.
Agentproof: Static Verification of Agent Workflow Graphs
arXiv preprint
This work presents Agentproof, an approach for the static verification of agent workflow graphs, enabling correctness properties of LLM-based agent workflows to be checked before execution.
Enabling dependable flexibility in industrial automation with formal methods integrated to development toolchains
PhD Dissertation, Luleå University of Technology
This thesis develops methods and toolchains that integrate formal verification into the engineering of flexible, reconfigurable industrial automation systems, enabling dependable flexibility across the development lifecycle of IEC 61499-based control software.