All work
Moving a pipeline onto a shared data-cache layer
Migrated an entire platform’s recommendation stack off direct report reads onto the shared data-cache architecture, then built new recommendations, audits and dashboard widgets on top.
- When
- Dec 2023 — Sep 2024
- Role
- Engineer
- Context
- Bringing Microsoft Advertising to parity with Google Ads
Context
The recommendation product started on one ad platform. The second platform’s side of the pipeline still read data the old way — straight from the provider’s reports — so it couldn’t share the cache, parsing and aggregation the main platform used.
What I built
- Migrated the full Microsoft Ads suggestion stack to the data-cache architecture — keyword bidding, first-page bids, ad groups without RSAs, losing-ad pruning, auto-apply and tidbits — plus report-parsing and metric-aggregation fixes.
- New Microsoft optimizations: campaign and shared-budget increase suggestions, predictive match type, and RSA suggestions.
- PPC Audit Score for Microsoft Ads (launched Feb 2024) on the account and all-accounts dashboards.
- Microsoft Ads dashboard sections with Top Elements, Performance Change and Change History widgets, plus shareable dashboards (launched Sep 2024).
- Microsoft Ads onboarding as part of the 2023 onboarding redesign.
Outcome
The second platform reached parity on the shared architecture, and the same chassis later carried two more.