J
Joshua Long
About
Detail
Cumming, Georgia, United States
Résumé
Jobs
verified_user
0% verified
Senior Machine Learning Engineer & Data ScientistDoorDashDec 2022 - Current (3 years 10 months)DoorDash's ML ecosystem spans search & discovery, marketplace intelligence, and ML platform enablement. I worked on both applied ML systems and integrated deeply with the evolving ML Platform, enabling scalable deep learning and GenAI workloads across training, orchestration, serving, and observability.
GenAI / LLM + Platform Work
• Built and deployed LLM-powered semantic search pipelines using GPT-4, Claude 3, and Sentence Transformers (E5-Large), improving long-tail query understanding across menus, grocery, and retail.
• Integrated with DoorDash's ML Platform components: used Fireworks for training, Metaflow for orchestration, and Serving V2 (Triton) for scalable, low-latency inference with model warmup and custom pre/post- Machine Learning & Data ScientistClover HealthAug 2019 - Nov 2022 (3 years 4 months)Transformer-Based Clinical NLP
• Applied ClinicalBERT / BioBERT embeddings to extract features from physician notes, improving chronic condition detection and chart-review automation.
• Fine-tuned transformer models with task-specific objectives, balancing recall improvements with explainability requirements for clinical review.
• Built an early clinical summarization prototype using encoder-decoder transformer architectures to assist care teams.
• Enhanced diagnosis and medication extraction using transformer-based sequence classification.
• Ensured reproducible training and evaluation workflows across claims, EHR, and NLP pipelines.
Core ML & Data Science Work
• Developed Risk Adjustment Factor (RAF), ch - Machine Learning EngineerAmazonJan 2017 - Aug 2019 (2 years 8 months)• Built high-volume DeepAR forecasting models improving demand predictions for volatile ASIN clusters.
• Designed anomaly detection pipelines monitoring fulfillment center throughput, inbound/outbound forecasting errors, and inventory discrepancies.
• Developed Spark + EMR data pipelines processing billions of weekly events for backtesting and feature generation.
• Contributed to early internal feature-store standardization, improving feature consistency across retail ML teams.
• Implemented monitoring and anomaly detection for both data quality and model outputs in production supply-chain systems.
• Developed dashboards and alerting for forecast accuracy and data freshness, enabling faster detection of model degrad - Data ScientistBigCommerceJul 2015 - Dec 2016 (1 year 6 months)• Built the company's first merchant churn prediction model using logistic regression + boosted trees.
• Conducted segmentation, lifecycle modeling, and revenue analysis for merchant behavior insights.
• Improved search relevance via metadata enrichment and embedding-similarity scoring.
• Designed ETL and analytics pipelines using Python + SQL + Airflow supporting dashboard and ML workloads.
Education
verified_user
0% verified
- M.S., Computer ScienceGeorgia Institute of TechnologyAug 2020 - Dec 2021 (1 year 5 months)
- B.S., Computer ScienceUniversity of GeorgiaAug 2011 - Jun 2015 (3 years 11 months)