Selected work

Systems built for the messy parts of production.

My work spans remote agents, cloud services, observability, update and recovery infrastructure, and high-reliability media software.

Environment

Edge ↔ Cloud

Primary constraint

Failure is normal

Method

Observe · Verify · Recover

01

Distributed systems · Current

Edge agent platform

GoSQLiteDistributed queuesLinux / Windows

Modernizing a long-lived remote agent into Go, with durable local queues, cloud command handling, heartbeats, scheduling, and behavior-preserving migration from a legacy runtime.

Engineering focus: Delivery semantics, restart safety, and clear ownership across concurrent workers.

02

Platform engineering · Current

Safe update and recovery system

GoA/B slotssystemdWindows SCM

Designing a self-update architecture for remotely deployed software: verified artifacts, A/B installations, atomic activation, candidate probation, crash-loop detection, and automatic rollback.

Engineering focus: The known-good version remains recoverable throughout the entire transaction.

03

Cloud & reliability · Current

End-to-end observability

OpenTelemetryOpenObserveKubernetesOTLP

Instrumenting agent workflows across callbacks, schedules, persistent queues, and cloud commands using traces, structured logs, and metrics—then validating them on real virtual machines.

Engineering focus: Trace context follows work across process, persistence, and network boundaries.

04

Media systems · Production

Cinema playback platform

RustwgpuGStreamerDCI / SMPTE

Building DCI/SMPTE-compliant playback infrastructure with secure media processing, custom pipeline components, and GPU-accelerated rendering for cinema environments.

Engineering focus: Predictable playback where a software failure becomes a dark screen for an audience.

05

Reliability & compliance · Production

Secure playback telemetry

RustBinary formatsXMLWeb tooling

Designed tamper-resistant playback records and human-readable diagnostic tooling for forensic investigation, operational support, and cinema compliance workflows.

Engineering focus: Operational evidence must remain trustworthy even when something goes wrong.

06

Systems architecture · Production

Native-library modernization

C / C++Rust FFIlibasdcpAgent skills

Creating safer language boundaries around mature C and C++ cinema libraries, while exploring repeatable workflows for behavior-preserving modernization.

Engineering focus: Reduce unsafe surface area without discarding proven domain behavior.

The common thread

Make difficult systems easier to reason about.

Languages and platforms change. The work remains: define contracts, expose failure, preserve evidence, and make recovery explicit.

More about me →