Case Study · Developer / Corporate Event
Twilio: live demos and registration that could not afford a dropped packet
A developer-facing corporate event where on-stage demos, registration, and production all depend on connectivity in front of a technical audience. LGL Networks ran the mission-critical network as one accountable operator, setup to teardown.
Context
Twilio hosted a developer-facing corporate event — and a developer audience raises the bar: live API demos, workshops, and on-stage code are watched by people who notice the instant something lags.
LGL Networks operated the event’s mission-critical network as a segmented Critical Operations Network overlay, behind the venue circuit — supplementing it where permitted and never bypassing venue exclusivity.
Challenge
The risk profile combined corporate-event stakes with developer-event intensity:
- Live, on-stage demos in front of a technical crowd. A stalled demo is not just an inconvenience — it undercuts the credibility of a developer platform with the exact audience that judges it.
- Revenue-critical systems with no single owner. Registration, badge printing, any onsite payments, and production were each owned by a different vendor, with no one accountable for the network as a whole.
- Mixed traffic, mixed trust levels. Production, registration, staff, sponsor, and attendee traffic on one fabric is a reliability and compliance-scope hazard.
- No documented proof before go-live. A high-visibility keynote needs verified resilience, not best-effort setup.
Approach
We followed the standard LGL motion — Risk Assessment, written venue authorization, Critical Operations Overlay, monitored through teardown.
Segmented overlay, wired-first
We deployed a segmented LAN with wired drops for the event- and revenue-critical systems — keynote/demo stage, registration, any payments, production, and staff operations — wired first for paths that cannot tolerate wireless contention. Our documented segmentation standard isolated production, registration, payment, attendee, contractor, and management traffic into distinct zones.
Connectivity resilience
Where venue policy permitted, we ran dual-WAN with defined failover; otherwise an approved single underlay with a documented failover plan and target window. LGL procures dedicated, redundant upstream circuits rather than reselling commodity bandwidth.
Pre-opening acceptance + failover testing
Before go-live, we ran the documented pre-opening acceptance test — confirming segmentation held, WAN failover landed inside its target window, critical-system redundancy was live, and monitoring was reporting. Demo and keynote paths were validated under load.
Onsite engineering + NOC monitoring
An onsite network lead owned floor incident response, backed by remote NOC monitoring from setup through teardown.
Outcome
The engagement closed with a post-event report documenting acceptance-test results, incidents and time-to-failover, and segmentation integrity. The measured results for this event — acceptance-test outcome, incident count, failover times, and keynote and demo-stage network availability — are documented in the client’s post-event report.