Shipping software since 2019. Currently piloting
Uptime Track at Whatfix — an autonomous SRE
that diagnoses incidents while the rest of the fleet sleeps.
Prior life: platform engineering. Built the
launchpads other engineers deploy from. Comfortable in the dark margins
where Kubernetes meets Java meets Python.
Shipping an agentic SRE that autonomously investigates
production incidents and returns a descriptive RCA with
concrete next steps. Less paging humans at 3 AM. More sleep for everyone.
STAGE 01 · CLEARED
SOFTWARE ENGINEER I
PRIOR
▸ Platform Engineering
Built abstraction frameworks so developers could ship to
Kubernetes without learning Kubernetes.
Scaled services across the platform layer and kept the deployment lanes
clear at peak traffic.
An autonomous agent that watches your production fleet, smells smoke
before the alerts fire, and ships a human-readable RCA
with the next-three-steps already drafted.
INVESTIGATE▸CORRELATE▸DIAGNOSE▸REPORT
uptime-track ▸ incident detected cluster=prod-eu-west · service=billing-api signal: p99 latency 4.2s ↑ (baseline 180ms)▸ correlating 12 telemetry streams …▸ hypothesis space ↦ 4 candidates▸ ROOT CAUSE (87%) redis pool exhaustion after deploy PR #2914 reduced pool size 200 → 50▸ NEXT STEPS 1. revert PR #2914 2. raise pool to 250 in staging 3. add pool-saturation SLO generated in 4.7s · agent v0.9.2▸ _
