Jules Elm

In February 2025 I hired and coached engineers while introducing peer design reviews. The engineering group adopted peer design reviews.

Jules sponsored the peer design-review pilot and coached three engineering managers. Engineers Mara Vale and Theo Reed designed the review format together. Hiring was shared with the managers. The engineering group adopted peer design reviews; we have not collected feedback about which coaching was most useful.

I sponsored the peer design-review pilot and coached three managers. Mara and Theo designed the format, and the managers shared hiring responsibility. Please keep that division of work explicit. I still cannot say which part of my coaching helped most.

2024-11: A supplier arrangement needed a confidential technical review. Give procurement a technical assessment without disclosing internal contract details. Coordinated engineering and procurement input and wrote the technical tradeoff assessment. Procurement used the assessment in its supplier review; commercial terms and the final decision remain private.

2023-05: Four product teams requested different platform improvements from the same small group. Agree priorities and ownership within the available engineering capacity. Worked with engineering managers and product leads to compare support burden and migration effort. The group agreed a two-quarter platform adoption plan and named an owner for each migration; revenue impact was not measured.

2021-02: Product and operations disagreed about replacing a shared service in one release. Make the migration tradeoffs explicit and agree a sequence the teams could support. Mapped failure boundaries with both groups and authored a staged separation proposal. The teams chose a staged separation and retained a rollback path; product teams implemented the changes.

In March 2019 I designed a deployment rehearsal format and co-built its runner with engineers. Two teams adopted rehearsals before deployment.

2017-10: Teams had recovery instructions but had not practised making decisions together. Help service owners and support staff rehearse escalation decisions. Facilitated four incident exercises with service owners and support staff. The group clarified escalation owners and recorded gaps in recovery instructions; outage reduction was not measured.

2016-03: Service teams could not distinguish a quiet queue from a stalled consumer. Expose queue age and ownership without replacing the messaging platform. Designed a queue-age dashboard and paired with service owners to set alert thresholds. Five services adopted the dashboard and ownership alerts.

The March 2016 queue-age dashboard and the "service health dashboard" in the resume are the same project. Jules designed the dashboard and worked with the five service owners on alert thresholds. This is not a second delivery.

2013-09: Routine recovery depended on an engineer remembering undocumented steps. Help the operations team own routine recovery safely. Paired with operators to write recovery runbooks and rehearse failure cases. Operations maintained the runbooks and performed routine recovery; no response-time baseline was recorded.

2012-04: Import failures were discovered after staff had started processing records. Make failures visible before processing and leave a recovery path for operators. Built validation checks and a rejected-record queue with the service-delivery team. Operators used the checks across 18 recurring data feeds.