I build the parts of a system that have to be right, not just fast β workflow engines, async data layers, streaming pipelines, and the compilers and schedulers underneath them.
![]() |
zigflow/zigflow: declarative YAML workflows on Temporal, with no SDK boilerplate. Zigflow compiles a workflow written in YAML (the CNCF Open Workflow Specification) into a full Temporal workflow, with retries, state management and deterministic execution. Teams get Temporal's durability without writing Go, Java or TypeScript workflow code. π οΈ Maintainer since October 2026: reviewing PRs, keeping CI and dependencies healthy, and driving the road to v1. |
Highlights
- #615: activity inputs resolved in the workflow. Every step now receives, and history records, the exact values the workflow computed, so a run can be debugged from its history alone.
- #623: unblocked the project's CI. A CVE in
golang.org/x/nethad failed the image scan onmainand on every open PR for two days. The fix brought builds and image publishing back to green. - #591: cancelling a workflow now stops it. A fork waits for its branches to be cancelled instead of returning while they keep running.
15 merged pull requests across 2 repositories, and 3 issues reported. Took part in 10 discussions. Reviewed 4 pull requests.
Latest merged:
- zigflow/zigflow#657: ci: run the pr review job on ubuntu-24.04 (2026-11-11)
- zigflow/zigflow#644: feat: add scheduleWorkflowType and deprecate scheduleWorkflowName (2026-11-09)
- zigflow/schema#22: feat: add scheduleWorkflowType and deprecate scheduleWorkflowName (2026-10-28)
Full record: CONTRIBUTIONS.md
- βοΈ Workflow orchestration & durable execution β Temporal, compensation, state you can reason about a week later
- π§© Compilers & DSLs β graph traversal, lowering, validation, normalisation; the interesting bits are all edge cases
- β‘ Async Python at the data layer β SQLAlchemy 2.0, connection lifecycle, migration discipline
- π Scheduling correctness β recurrence rules, occurrence generation, timezone-aware scheduling
- π Auth & access control β asymmetric JWT, centralised verification, RBAC that holds at the fifth service
- π₯ Real-time media β RTSP, HLS, WebRTC, and the bandwidth arithmetic behind choosing between them
- π Observability of distributed work β tracing an execution back to the intent that produced it
Languages & Frameworks
Orchestration & Data
Infrastructure
Media & Protocols
- π Deep in workflow orchestration β compilers, durable execution, making failures legible
- πΉ Learning Go properly, mostly by fixing things in a runtime written in it
- π Contributing upstream instead of carrying patches locally
- βοΈ Getting better at the non-code half β design ownership, root-cause analysis, and writing things down
π Python backend & distributed systems Β· βοΈ Workflow orchestration & Temporal Β· π§© Compilers, DSLs and developer tooling Β· π Open source



