- MLOps North
- Toronto Summit
- Presented by TMLS
Cost Management and ROI
About the track
Uber spent their 2026 AI budget in four months, OpenAI cut Luna prices by 80% at the end of July, and open source models are being used much more due to exploding costs. AI usage is continuing to grow, yet leaders are struggling to quantify the ROI on their multi-billion dollar investment. This track is for engineering managers and above who are managing budgets for AI.
Talks cover the types of AI budgets for different kinds of projects (productivity, cost reduction, process transformation), debugging and optimizing your AI costs, and demystifying the cost consoles in hyperscalers and model providers. Speakers are finance and technology leaders who’ve optimized AI costs. You’ll leave knowing how to create an ROI plan for your projects, plus a checklist of model optimizations.
Track host
SVP AI & Operations, Wisedocs
See Track Lead Profile
Denys Linkov is SVP of AI at Wisedocs, where he leads the company’s AI strategy and delivery for medical document intelligence, and a machine learning lecturer at the University of Toronto. Previously he was Head of ML at Voiceflow, guiding enterprise AI initiatives and working with more than 50 enterprises on their conversational and generative AI journeys, following earlier engineering work at LinkedIn.
Denys’s experience spans the full AI product lifecycle, from building production ML systems and evaluation frameworks (including his widely shared work on micro-metrics for LLM evaluation) to directing product teams and advising customers on best practices. He’s also a prolific educator: his courses and generative AI content have helped over 150,000 learners build practical AI skills, and he’s a regular speaker at conferences including QCon, AI Engineer, and TMLS.
Toronto Summit · Presented by TMLS · Nov 5–6, 2026 · RBC WaterPark Place, TorontoThe people building agents, and the people building with them – one room, on the Toronto waterfront. Two days of practitioner-curated talks on what actually ships: the stack underneath agentic systems, and the products, workflows, and agents teams are running in production right now.