Olivier Pomel
Co-founder-CEO of Datadog
Co-founder-CEO of the monitoring giant Datadog, who has pushed the company into agentic AI — shipping autonomous “Bits AI” agents that investigate incidents — while arguing observability is what will make AI systems trustworthy.
In his words
By the time customers wake up, there is already a hypothesis of what is wrong.
Trust requires accuracy well above 90%. While humans can be effective at 70% accuracy, AI systems need much higher standards to earn adoption.
Biography
Olivier Pomel is a French software engineer who co-founded Datadog in New York in 2010 with Alexis Lê-Quôc, building it into one of the dominant cloud-monitoring and observability platforms and taking it public in 2019. His engineering roots are in systems and infrastructure — the plumbing that keeps large software running.
Betting Datadog on agents
Under Pomel, Datadog has moved aggressively into AI, releasing domain-specific agents — Bits AI SRE, Bits AI Dev Agent, and Bits AI Security Analyst — that autonomously investigate incidents, propose fixes, and triage alerts. Pomel argues the biggest productivity gains come not from flashy generation but from automating firefighting: incident response, false alarms, and system maintenance. Datadog has also positioned itself to monitor its customers’ AI systems, offering “LLM observability” for the new agent-driven stack.
Pomel is notably measured about the bar AI must clear. He stresses that trust demands accuracy well above 90% — far higher than what humans get away with — because operators will not hand production systems to an agent they cannot audit. It is a builder’s optimism tempered by an operator’s caution: AI as a second pair of eyes that must earn its keep before it earns the keys.
Where they stand in the war
Who backs them up
Sources & further reading
Canonical record: https://battlelines.ai/topic/olivier-pomel