Ongoing
Autonomous infrastructure: how far can self-healing go?
An investigation into closed-loop infrastructure automation — from auto-remediation to AI-driven operations — and where human accountability must survive.
Research
Research
Long-horizon technical investigations, written up with context, method, findings, and open questions. Research here is practitioner-driven: it starts from real enterprise and lab problems, not from citation counts.
Active
Datacenter energy as the binding constraint on computing
Why power availability — not chips or capital — is becoming the limiting factor for AI-era infrastructure, and what it means for architecture decisions today.
Active
Measuring AI Integrity: Toward Useful Metrics
An active research program on measuring AI integrity: why accuracy metrics miss it, dimensions like grounding faithfulness, and harness design.