Voice AI Connect Sydney: Optimising Token Spend for Engineering
Voice Connect Sydney: Optimising Token Spend for Engineering
Your last AI invoice surprised you. This session makes sure the next one doesn't.
Every engineering team shipping AI is staring at the same line item: tokens. Burned by untuned prompts, model choices made on vibes, and spend nobody can trace. Most teams can't answer two simple questions - what did AI cost us last month, and what did it buy us?
In 90 minutes, we'll fix that. We start with lightning talks from Telnyx partners - real teams showing production voice AI they've shipped on our stack. Then our Global VP of Sales, Alex Sobin, walks through the token spend playbook with the pricing table on screen: model choice (the same task can cost 20x less on input tokens depending on the model), cached input at up to 88% off, service tiers matched to latency needs, and governance that makes spend visible per team before it happens.
What you'll leave with:
A cost model you can apply to your own AI stack the next morning
The real pricing table across open-weight models - and when each one wins
This event is for:
Engineers, ML engineers, engineering managers, platform teams and technical founders - anyone building AI features or owning an AI budget. Voice AI builders welcome; no voice AI experience required.
Event details:
Date: Wednesday 21 October
Time: 12PM - 1:30PM
Location: ACS Harbour City Labs, Barangaroo
Lunch is on us. No prep needed - just bring your questions.
Space is limited - all registrations are subject to approval. You'll receive confirmation within 48 hours.