Toni Anev

Case studies

Selected work

Public summaries of platform and ML engineering work. Details are condensed, but the scope and outcomes are representative.

Sanofi · Machine Learning Engineering Lead

AI infrastructure and ML platform scale-out

Led platform architecture and delivery systems for AI/ML workloads used across global R&D teams, with an emphasis on throughput, reproducibility, and production readiness.

  • Scaled a global AI / HPC platform spanning Nvidia A100/H100, EKS, FSx, and S3-backed storage patterns.
  • Standardized Terraform-based IaC and automated compute provisioning to improve platform throughput by 23%.
  • Operationalized internal models through SageMaker and Bedrock, reducing time-to-production by 26% across 7+ teams.

Stack · AWS · EKS · FSx · S3 · SageMaker · Bedrock · Terraform · HPC

Petco · Technology Lead

Pricing and personalization ML delivery

Led ML engineering delivery for pricing and personalization systems, modernizing platform standards and improving the speed of experimentation and deployment.

  • Built and led a 5-person ML engineering team and introduced modern MLOps practices on GCP.
  • Redesigned a price-elasticity platform, reducing runtime from 24 hours to 4 hours and compute load by 32%.
  • Directed a personalization program that delivered a 3% basket-size uplift and approximately $13M/month in incremental revenue.

Stack · GCP · GKE · Dataproc · Vertex AI · Airflow · Snowflake · Seldon

Loblaw Companies Limited · Data / MLOps Engineering

Enterprise pricing, promotions, and segmentation systems

Built and scaled production data / ML systems across pricing, promotions, and targeted marketing, from early decisioning engines to enterprise MLOps orchestration.

  • Led rollout of a promotion optimization engine across 16 divisions and 200+ product categories on GCP, Terraform, and Airflow.
  • Built orchestration services that reduced batch latency by 73% and lowered compute cost by $2.7K/month.
  • Contributed to segmentation and pricing systems tied to $42M+ annual incremental revenue and $15–20M revenue lift in earlier pricing decisioning work.

Stack · GCP · Dataproc · Cloud Composer · Terraform · Airflow · Microservices

Notes

Public summaries only

Some implementation detail is confidential. I’m happy to discuss architecture decisions, tradeoffs, and delivery patterns in a direct conversation.

Contact Proof artifacts Open CV