| |
|
| |
| DETAILS |
|
The AI Infrastructure Trade-off: Building Production Stacks That Scale Profitably
VentureBeat WEKA | New York City
You've deployed AI. Now comes the harder part: infrastructure that scales economically, orchestration layers that don't break under load, & data pipelines that cost less than the models they feed.
This private AI Impact salon brings together infrastructure leaders who've moved past deployment & are now optimizing for the long game - making architectural choices that hold up when model capabilities, pricing, & tooling change quarterly.
What You'll Gain
Production infrastructure economics - How Wonder manages AI/ML costs across real-time ordering, logistics, & multi-brand platforms where latency SLAs meet budget constraints.
Agent orchestration models that pencil out - WEKA's Val Bercovici on routing strategies, token economics across autonomous workflows, & the infrastructure primitives that differentiate performant from prohibitively expensive agentic systems.
Cost-performance levers that actually move the needle - Storage & memory choices that reduce both latency & spend, eval frameworks that matter in production, & where to build versus buy over the next 12-24 months.
Featured Sessions
Building Infrastructure That Survives the Next 18 Months
James Chen, Chief Technology Officer, Wonder
James oversees platform architecture spanning Wonder's family of brands-Grubhub, Blue Apron, Tastemade-where real-time ops meet data platforms & production AI/ML. Expect candid discussion on cost/SLA trade-offs, reliability patterns for multi-workload environments, & hard-won lessons from integrating systems & teams across acquisitions.
Agent Economics: Routing, Memory, & the Real Cost of Autonomy
Val Bercovici, Chief AI Officer, WEKA
As enterprises shift from single-model inference to multi-agent orchestration, the cost model changes fundamentally. Val explores model routing decisions, token spend across autonomous workflows, memory & storage architecture for agent systems, & infrastructure choices that determine whether your agentic deployments scale profitably or blow up your budget.
Why Attend
Trade playbooks with peers making similar infrastructure decisions-not vendor pitches, but the practical detail that helps you choose architectures, set SLAs, & allocate budget wisely.
See where the efficiency battles are being won-storage choices that cut both latency & cost, orchestration patterns that scale without exploding spend, & the vendor/build decisions that differentiate sustainable AI operations.
Leave with an actionable plan for the next 12 months-what infrastructure investments deliver ROI, where the real leverage points are, & which architectural choices won't require rewrites when the ecosystem shifts.
Seats are limited. Attendance is subject to approval.
|
|
|
|
|
|
|
|
