Company

LinkedIn

36 cases from LinkedIn (Social, SaaS B2B, Marketplace), each digested from a public talk or engineering post and linked to the original. 12 were checked in an online A/B test; 15 report results in production.

Tasks: Retrieve and rank from millions of items, Build a data platform, Rank search results, Evaluate models, Catch fraud and abuse and 15 more. Methods: Human in the loop, Two-tower model, Embedding fine-tuning, LLM-as-a-judge, A/B test, Dense retrieval and 92 more. Tools: Apache Spark, Apache Kafka, Ray, TensorFlow, Apache Hadoop, XGBoost and 29 more.

Tech blog: linkedin.com

Inside LinkedIn's data context layer: curating trustworthy context for humans and agentssocial2026-10
Rebuilding LinkedIn's Follows Recommendations with LLM-Based Semantic Retrieval and Rankingsocial2026-08
Semantic Search for AI Agents at Scale: Retrieval and Ranking for LinkedIn’s Hiring Assistantsocial2026-06
Crosscheck: Benchmarking AI models in the real worldsocial2026-05
LinkedIn Cognitive Memory Agent: long-term memory for personalizing LLM agentssaas-b2b2026-03
Building the incremental and online training platform at LinkedInsocial2025-10
Building the next generation of job search at LinkedInmarketplace2025-05
JUDE: LLM-based representation learning for LinkedIn job recommendationsmarketplace2025-05
Buyer Journey Insights with Data-Driven Attributionsocial2025-01
Automated GenAI-driven search quality evaluationsocial2024-12
Practical Text-to-SQL for Data Analyticssocial2024-12
How LinkedIn Audience Network protects advertisers from invalid trafficadtech2024-08
Candidate Generation in a Large Scale Graph Recommendation System: People You May Knowsocial2024-06
Matching LinkedIn members with the right Premium productssocial2024-05
Group’s anatomy: Analyzing your LinkedIn Groups with real-time insightssocial2024-05
Evaluating the Success of Consumer Generative AI Productssocial2024-03
Building a Large-Scale Recommendation System: People You May Knowsocial2024-02
Privacy preserving single post analyticssocial2023-12
Augmenting our content moderation efforts through machine learning and dynamic content prioritizationsocial2023-11
How LinkedIn Is Using Embeddings to Up Its Match Game for Job Seekerssocial2023-10
New Approaches For Detecting AI-Generated Profile Photossocial2023-06
Viral spam content detection at LinkedInsocial2023-04
Building and maintaining the skills taxonomy that powers LinkedIn's Skills Graphsocial2023-03
Improving the customer’s experience via ML-driven payment routingfintech2023-01
Ocelot: Scaling observational causal inference at LinkedInsocial2022-12
Challenges and practical lessons from building a deep-learning-based ads CTR prediction modeladtech2022-08
Applying multitask learning to AI models at LinkedInsocial2022-08
Measuring downstream impact on social networks by using an attribution frameworksocial2022-07
Measuring marketing incremental impacts beyond last click attributionsocial2022-07
Improving job matching with machine-learned activity featuresmarketplace2022-05
Community building recommender for group chats in LinkedIn Messagingsocial2022-04
Matching external companies to LinkedIn's Economic Graph at scalesocial2022-04
The journey to build an explainable AI-driven recommendation system to help scale sales efficiency across LinkedInsaas-b2b2022-04
Near real-time features for near real-time personalizationsocial2022-03
Using Bayesian optimization for balancing metrics in recommendation systemssocial2022-02
How our content abuse defense systems work to keep members safesocial2022-01