Skip to content
View FTVarna's full-sized avatar
  • London

Block or report FTVarna

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
FTVarna/README.md

Fevzi Tugrul Varna, PhD

Lead AI Engineer and Observability Architect
Visiting Research Fellow in Informatics, University of Sussex · Co-founder, SwarmLab

I build C# and .NET systems for warehouse operations and business workflows. A question that guides my work is: when one service reports success but the next never completes its task, what evidence would let an engineer explain why?

Software engineering

  • Observability for distributed systems. I lead development of a platform that brings together logs, metrics and traces from warehouse management software, material flow controllers and RabbitMQ. I use OpenTelemetry and Grafana to help engineers follow an operation across application boundaries.
  • AI-assisted root cause analysis. I design investigations that give a model incident telemetry and a map of application dependencies. I evaluate whether the evidence supports each finding, what the model missed, and how much time and cost the investigation required.
  • Business workflow automation. I built .NET services for invoice intake, supplier verification and document extraction, with dashboards that let operators review the results.

My day-to-day stack includes ASP.NET Core, EF Core, SQL, RabbitMQ and Azure. My deployment work spans Azure Container Apps, Kubernetes and OpenShift.

Selected public work

My research with Professor Phil Husbands explores how a swarm can share useful discoveries without every agent abandoning its own search too early.

Work What to look for
HIDMS-PSO · MATLAB implementation How groups of particles use different search behaviours and exchange information as the optimisation runs.
AHPSO and BEPSO · journal article · BEPSO code Two ways for agents to cooperate: lending energy to another particle, or learning when to act on a signal intended for another group. I developed the software and ran the experiments for this study.
E-AHPSO · Applied Soft Computing How combining an exploratory swarm with methods that refine promising solutions changes performance across benchmark problems.

Full publication record · ORCID

@FTVarna's activity is private