When the Agent Becomes the User: Rethinking Observability for Autonomous Software
About
Adam Hicks
Adam Hicks is a Senior Solutions Architect and technologist at groundcover, where he helps enterprise teams adopt eBPF-based, bring-your-own-cloud observability for Kubernetes and AI workloads. He's spent more than a decade in the observability and cloud-native space, including as a Solutions Architect at Honeycomb.io, as Field CTO and Practice Director at Trace3, where he built the firm's observability and platform engineering practices, and as an early engineer at Morpheus Data through its acquisition by HPE. Adam is known for his deep OpenTelemetry expertise, and he works hands-on with teams on distributed tracing, GenAI observability, and moving from proprietary instrumentation to open standards. When he's not deep in telemetry pipelines, you'll find him biking and skiing near his home in the White Mountains of New Hampshire.
In this episode, host Logan Lyles sits down with Adam Hicks (Senior Solutions Architect, Groundcover) and Justin Billig (CEO & Co-founder, Tensure) to explore what happens when AI agents become users of enterprise platforms — investigating production incidents, consuming observability data, and increasingly taking action inside complex software environments.
The conversation examines why traditional observability tools aren't necessarily designed for AI agents, how connected operational context can improve agent effectiveness, and what financial services organizations need to consider as they balance greater autonomy with security, compliance, cost, and reliability.
The conversation covers:
- Why traditional logs, metrics, traces, and dashboards aren't enough when AI agents become the consumers of observability data
- How knowledge graphs and connected telemetry help agents understand relationships across production systems
- Who is accountable when autonomous agents make changes — and why identity, permissions, and auditability matter
- How Riskified consolidated fragmented observability tools to improve developer workflows and agent access to production context
- What Nimbus learned about balancing telemetry fidelity, data privacy, regulatory requirements, and observability costs
- How Groundcover uses bring-your-own-cloud architecture and eBPF to support context-rich observability
- What CIOs, CTOs, and CISOs should evaluate before giving AI agents access to sensitive production information
- Why internal developer platforms are evolving from developer portals into control layers for humans and agents
- How self-service and golden paths can prevent platform engineering teams from becoming bottlenecks
- Why production performance, testing feedback, rollback, and recoverability are increasingly important as agents become more autonomous
- How Tensure and Groundcover are partnering to help financial services organizations prepare for agentic engineering
Smooth shipping is a few steps away
Let's see how we can help your team move faster. From developer platforms to cloud infrastructure and AI solutions that get your developers shipping again.