No Shortcuts: Making Agentic Network Operations Reliable Enough to Trust – CodiLime Triple-T
The industry’s current answer to every network-operations problem is “put an LLM on it.” And then the LLM chokes on a million rows of telemetry, reasons differently about the same incident twice, and takes a different action on Tuesday than it did on Monday. That is not a model problem — it is an engineering problem, and it has an engineering answer. In this session CodiLime shows how to build an agent that operators can genuinely trust: treat the LLM as an intelligent abstraction layer over deterministic tools, proven statistics and classic ML, solid telemetry and tested automation — and refuse every tempting shortcut. The proof is a working predictive-troubleshooting agent that sees a failure hours before it happens, names the cause, and proposes the fix.
Tomasz Janaszka is a solutions architect with over 20 years of experience in the telco/IT industry. He is a doctor of technical sciences in the field of telecommunications and an experienced engineer who has worked for many years in research and operational projects. He has worked as a leader, as a solution architect and as a developer in projects covering networks design, capacity planning, traffic engineering, load balancing, resources optimization and focusing on software solutions supporting automation of business processes related to network management. Currently working in CodiLime’s R&D department, he explores AI/ML technology looking for practical applications in various networking aspects. Tomasz is the author and the co–author of several scientific publications and a speaker at conferences, webinars and technical workshops. Check out his ebook “AI and Machine Learning for Networks” : https://resources.codilime.com/ai-and-ml-for-networks
