Looking for a job

Dennis Scott

@dennisscott

Message

Senior machine learning engineer building scalable real-time ML systems that lift CTR, conversion, and reliability.

United States

Message

What I'm looking for

I’m looking to lead end-to-end ML and real-time inference—turning complex business problems into reliable, low-latency systems through strong experimentation, MLOps, and observability.

I’m a Senior Machine Learning Engineer with 9+ years building and scaling production ML systems across ads ranking, pricing, AI safety, and personalization platforms.

I’ve delivered measurable impact at companies like Facebook (Meta), Opendoor, Robust Intelligence, and Attentive—improving CTR, conversion rates, model reliability, and system scalability while balancing latency, scale, and reliability.

My work is end-to-end: recommender and ranking models, real-time inference pipelines, feature engineering, and experimentation. At Attentive, I deployed two-tower retrieval plus embedding-based ranking and ANN/vector search to drive a 12% CTR lift across high-volume campaigns, and I scaled distributed inference to enable sub-second decisions for 5M+ users per send under strict latency SLAs.

I’m also deeply focused on modern AI technologies—LLMs, RAG, and agentic workflows—paired with strong MLOps and observability. I’ve replatformed ML workflows (training time 10h → 3h), reduced feature latency from hours to 30 minutes, and strengthened LLM safety by driving unsafe outputs down to ~1% in production.

Experience

Work history, roles, and key accomplishments

Current

Senior Machine Learning Engineer

Current

Attentive

Jul 2024 - Present (1 year 11 months)

Built and productionized a real-time personalization AI platform across SMS, email, and push, achieving a 12% CTR lift via two-tower retrieval, embedding-based ranking, and ANN/vector search. Scaled distributed inference to meet sub-second latency SLAs for 5M+ users and reduced ML training time from 10h to 3h.

Python PyTorch TensorFlow Kubernetes Ray Airflow Feature Stores (Feast)Vector Search Kafka

Machine Learning Engineer

Robust Intelligence

Jan 2023 - Apr 2024 (1 year 3 months)

Engineered runtime AI security for an AI Firewall, reducing false positives by 18% using hybrid Unicode/regex/YARA + ML detection pipelines and bringing mitigation turnaround to 24 hours. Deployed low-latency inference services (p95 80ms) and reduced unsafe outputs to 1% with semantic similarity and LLM evaluation.

Python PyTorch fastAPI Kubernetes XGBoost LightGBM LLM Evaluation Datadog

Machine Learning Engineer

Opendoor

Oct 2020 - Jul 2022 (1 year 9 months)

Enhanced large-scale valuation models to produce price and uncertainty estimates for real-time acquisition decisions, improving accuracy with comparable-sales modeling, geospatial embeddings, and feature engineering. Built ingestion pipelines that reduced onboarding time to 48 hours across 10+ markets and reduced overconfidence by 25% while cutting manual reviews by 30%.

PyTorch TensorFlow Dask Airflow Kafka AWS S3 Quantile Regression Bayesian Calibration MLFlow

Machine Learning Engineer

Facebook

Mar 2016 - Oct 2020 (4 years 7 months)

Improved large-scale ads ranking by building high-impact feature pipelines (CTR/CVR and contextual/user interaction signals), driving 0.6% CTR and 0.4% CVR lift. Optimized real-time training and inference (feedback latency to 15 minutes; p95 latency down 18% with 25% higher throughput) using stream processing and production serving systems.