Software Engineering Daily Engineering AI Systems for Autonomy and Resilience with Krishna Sai
63 snips
Feb 24, 2026 
Krishna Sai, CTO at SolarWinds and veteran of enterprise IT and observability, discusses designing AI-driven, agentic systems for resilient operations. He talks about moving from monitoring to ambient agents, architecting LLM gateways and safe abstractions, compressing and summarizing operational data, and balancing autonomy with human-in-the-loop controls.
AI Snips
Chapters
Transcript
Episode notes
Preprocess Logs Before Sending To Models
- Do not feed raw logs directly into LLMs; compress, deduplicate, and summarize first.
- Krishna Sai warns raw log firehoses are noisy and costly, so create compact representations of what changed before reasoning layers consume them.
Ticket Agent That Summarizes History And Suggests Replies
- An ITSM agent auto-summarizes incoming tickets using past incident history and suggests a response for quick human edit.
- Krishna Sai reports this instant adoption reduced MTTR and improved time-to-response because it fit existing workflows.
Embedded Agents Deliver Fast Measurable ROI
- Well-designed agentic features show immediate ROI when embedded in natural workflows.
- Krishna Sai cites examples with instant take rate and MTTR improvements of 30–50% for well-scoped use cases.
