Camilo Andrés Cáceres Flórez

Camilo Andrés Cáceres Flórez

Staff ML Engineer (Applied AI), Data Science Expert, and AI Architect leading GenAI, data science, and MLOps at scale. Ph.D. in Mechanical Engineering (Mechatronics) from UNICAMP and EMBA in Strategic Leadership, with experience driving end-to-end ML platforms, computer vision, and data science initiatives.

Auction and pacing simulations for ads lift

1 minute read

Published:

Simulations make lift estimates cheaper by testing policies before we spend budget. They are especially useful when experimentation cycles are long or randomized control is limited.

What to simulate

  • Auction outcomes: bid distributions, win rates, and clearing prices under new models.
  • Pacing and budgets: how new policies change spend velocity and delivery profiles.
  • Creative selection: whether creative policies skew delivery to certain cohorts or geos.

How to keep simulations honest

  • Replay production traffic with realistic throttling and consent-aware analytics.
  • Compare counterfactuals against holdout policies; calibrate against recent experiments.
  • Capture guardrails (CPC, CPA, ROI) with the same thresholds used in rollout gates.

Continue the conversation

Need a sounding board for ML, GenAI, or measurement decisions? Reach out or follow along with new playbooks.

Contact Subscribe via RSS or email See a case study

Leave a Comment

Related posts

Geo experiments for ads lift without slowing delivery

1 minute read

Published:

Geo experiments remain the most practical lift tool when auctions and budgets complicate randomization. The key is to design them so teams can run and ship without blocking roadmaps.

Experimentos geo para lift de anuncios sin frenar la entrega

1 minute read

Published:

Los experimentos geo siguen siendo la herramienta más práctica para lift cuando subastas y presupuestos complican la aleatorización. El objetivo es que los equipos puedan ejecutarlos sin bloquear el roadmap.

Ads ML as a subtopic of production ML systems

1 minute read

Published:

Ads ML shares the same control-plane skeleton as any production ML system; the difference is in the constraints. Bidding and pacing layers add strict latency and budget limits, but they still benefit from the same contracts, rollouts, and observability defaults.