Michael Alidi

Senior Machine Learning Engineer at HubSpot

Built the HubSpot AI platform (Content Assistant + ChatSpot) leveraging OpenAI GPT-4, DALL·E, and Breeze AI orchestration to deliver generative capabilities adopted by over 160,000 enterprise users across HubSpot CRM.

Machine Learning Engineer at HubSpot

Developed HubSpot's Ads Attribution engine using Bayesian regression, uplift modeling, and TensorFlow to attribute revenue across campaigns, enabling cross-channel attribution and data-driven marketing decisions.

Data Scientist at Google

Led data science efforts for unified DoubleClick/Ad Exchange analytics using TensorFlow, BigQuery, and Dataflow; consolidated ad performance measurement across 10B+ monthly impressions and multiple data streams within Google Marketing Platform.

Machine Learning Engineer at HubSpot

Developed HubSpot's Ads Attribution engine using Bayesian regression, uplift modeling, and TensorFlow to assign revenue credit across campaigns; built cross-channel ROI pipelines with TFX, Airflow, and BigQuery ML; deployed forecast models on Breeze AI + TensorFlow + Prophet to power monthly ROI projections across $250M+ in tracked ad spend with ±6% margin of error; built lookalike recommender using KMeans clustering + cosine embeddings on behavioral vectors from CRM graph, improving targeting across 4.5M user profiles; surfaced multi-touch attribution scores via Looker and Data Studio; partnered with data engineers, CRM architects, and AdTech PMs to co-design the Ads Attribution engine, aligning modeling assumptions with CRM identity graphs and multi-platform data from Meta, Google, and LinkedIn.

Senior Data Scientist at Google

Led Unified DoubleClick for Publishers and Ad Exchange using TensorFlow + BigQuery + Dataflow, enabling a consolidated ad-serving platform with real-time yield insights for thousands of publishers. Built forecasting pipelines with TensorFlow Extended (TFX) + BigQuery ML to predict inventory and optimize pacing across 10+ TB/day of auction data. Deployed anomaly-detection models with SciPy and Pandas on Cloud ML Engine, identifying delivery spikes and revenue leakage in minutes across billions of events per day.

Data Scientist at Google Marketing Platform

Contributed to Google Marketing Platform by integrating DoubleClick, Google Analytics 360, and AdWords data streams using BigQuery, Python, and Drémel SQL to unify ad performance across 10B+ monthly impressions.