Practical Resources

1. Specific Topics:

• [github.io] Self-Supervised Representation Learning

• [github.io] Contrastive Representation Learning

• [github.io] The Multi-Armed Bandit Problem and Its Solutions

• [ruder.io] An overview of proxy-label approaches for semi-supervised learning

• [mit.edu] Data-Centric AI vs. Model-Centric AI

• [towardsdatascience.com] Two-Tower Networks and Negative Sampling in Recommender Systems

• [towardsdatascience.com] Focal Loss : A better alternative for Cross-Entropy

• [fennel.ai] Real World Recommendation System – Part 1

• [fennel.ai] Real World Recommendation Systems – Part 2 (Training Data Generation)

• [fennel.ai] Real World Recommendation Systems – Part 3 (Modeling)

• [capitalone.com] Understanding TF-IDF for Machine Learning

• [fennel.ai] Feature Engineering for Recommendation System Series

• [fennel.ai] Feature Logging

• [fennel.ai] Information Leakage

• [towardsdatascience] What’s WRONG with Metrics?

• [elastic.co] Practical BM25 — Part 2: The BM25 Algorithm and its variables

2. Collection of Topics:

• [docs.aws] Building a Machine Learning Application

• [github.io] Lil’Log

• [ruder.io] A blog about natural language processing and machine learning

• [reachsumit.com] Sumit’s Diary

• [roizner.medium.com] Michael Roizner

• [amazon.science] Amazon Science: E-Commerce

• [youtube.com] MIT: Introduction to Data-Centric AI

• [fennel.ai] Fennel Blog

• [eugeneyan.com] Tech Blog

• [github.com] A Guide for Machine Learning Technical Interviews (FAANG Companies)

3. Videos

• [youtube.com] Upcoming Book: “Machine Learning System Design” by Alex Xu

• [youtube.com] Machine Learning (ML) System Design Interview Questions

4. Tech Blogs

• [engineering.fb.com] ML Applications

• [ai.meta.com] The latest AI news from Meta

• [engineering.fb] Core Infra

• [medium.com] AirBnB: AI & Machine Learning

• [netflixtechblog.com] Netflix TechBlog

• [blog.google] AI

• [engineering.atspotify.com] Spotify Engineering

• [ngineering.atspotify.com] Spotify Machine Learning

• [medium.com] Pinterest Engineering

• [eng.snap.com] Snapchat Engineering Blog

• [blogs.bing] Bing Blogs - The news, trends, people and technology behind Bing

• [evidentlyai.com] ML and LLM system design: 500 case studies to learn from

4. Case Studies

Year	Domain	Company	Topic	Article Link
2024	Social	Meta	Personalisation	Sequence learning: A paradigm shift for personalized ads recommendations
2024	Social	Meta	Infrastructure	Meta Andromeda: Supercharging Advantage+ automation with the next-gen personalized ads retrieval engine
2024	Social	Meta	Infrastructure	Meta’s approach to machine learning prediction robustness
2019	Social	Meta	Content Recommendation	Powered by AI: Instagram’s Explore recommender system
2023	Social	Meta	Scaling	Scaling the Instagram Explore recommendations system
2021	Social	Meta	Content Recommendation	How machine learning powers Facebook’s News Feed ranking algorithm
2024	Social	Meta	Robustness	Meta’s approach to machine learning prediction robustness
2023	Social	Meta	Content Recommendation	The AI behind unconnected content recommendations on Facebook and Instagram
2023	Social	Meta	Ranking	Our approach to explaining ranking
2023	Social	Meta	Ecosystem	How Meta measures the management of its AI ecosystem
2022	Social	Meta	Content Recommendation	The new AI-powered feature designed to improve Feed for everyone
2022	Social	Meta	Ads	New AI advancements drive Meta’s ads system performance and efficiency
2021	Social	Meta	Integrity	Harmful content can evolve quickly. Our new AI system adapts to tackle it
2020	Social	Meta	Integrity	Here’s how we’re using AI to help detect misinformation
2020	Social	Meta	Integrity	Training AI to detect hate speech in the real world
2020	Social	Meta	Integrity	How Facebook uses super-efficient AI models to detect hate speech
2020	Social	Meta	Integrity	How AI is getting better at detecting hate speech
2023	Social	Meta	Logging	Logarithm: A logging engine for AI training workflows and services
2023	Social	Meta	Instagram	Scaling the Instagram Explore recommendations system
2022	Social	Meta	Causal Inference	Improving Instagram notification management with machine learning and causal inference
2022	Social	Meta	Privacy	Applying federated learning to protect data on mobile devices
2022	Social	Meta	Data Engineering	Scaling data ingestion for machine learning training at Meta
2024	Social	Meta	Inference Framework	Taming the tail utilization of ads inference at Meta scale
2021	Social	Meta	Newsfeed Ranking	How machine learning powers Facebook’s News Feed ranking algorithm
2021	Social	Meta	Video	How Facebook encodes your videos
2020	Social	Meta	Diversity	On the value of diversified recommendations
2019	Social	Twitter	Feedback	Addressing Delayed Feedback for Continuous Training with Neural Networks in CTR prediction
2023	Chat	Snapchat	Retrieval	Embedding-based Retrieval with Two-Tower Models in Spotlight
2022	Chat	Snapchat	Privacy	Differential Privacy at Snapchat
2022	Chat	Snapchat	Feature Engineering	Synthetic Data for Machine Learning (ML)
2022	Chat	Snapchat	Feature Infra	Speed Up Feature Engineering for Recommendation Systems
2022	Chat	Snapchat	Ranking	Machine Learning for Snapchat Ad Ranking
2022	Chat	Snapchat	Video	Snap Video Compression
2022	Chat	Snapchat	Metrics	STOMP out Bad Metrics
2025	Image	Pinterest	Retrieval	Advancements in Embedding-Based Retrieval at Pinterest Homefeed
2025	Image	Pinterest	Retrieval	Establishing a Large Scale Learned Retrieval System at Pinterest
2024	Image	Pinterest	GNN	LinkSage: GNN-based Pinterest Off-site Content Understanding
2023	Image	Pinterest	Diversity	Building for Inclusivity: The Technical Blueprint of Pinterest’s Multidimensional Diversification
2023	Image	Pinterest	Ranking	Deep Multi-task Learning and Real-time Personalization for Closeup Recommendations
2023	Image	Pinterest	Ranking	Training Foundation Improvements for Closeup Recommendation Ranker
2023	Image	Pinterest	Diversity	Representation online matters: practical end-to-end diversification in search and recommender systems
2023	Image	Pinterest	User Understanding	Large-scale User Sequences at Pinterest
2022	Image	Pinterest	Real Time Personalization	How Pinterest Leverages Realtime User Actions in Recommendation to Boost Homefeed Engagement Volume
2024	Video	Netflix	Observability	Part 1: Title Launch Observability at Netflix Scale
2024	Video	Netflix	Observability	Part 2: Title Launch Observability at Netflix Scale
2024	Video	Netflix	Long Term Satisfaction	Recommending for Long-Term Member Satisfaction at Netflix
2024	Video	Netflix	Metrics	Improve Your Next Experiment by Learning Better Proxy Metrics From Past Experiments
2024	Video	Netflix	Video	Video annotator: a framework for efficiently building video classifiers using vision-language models and active learning
2022	Video	Netflix	Causal Inference	A Survey of Causal Inference Applications at Netflix
2024	Video	Netflix	Causal Inference	Round 2: A Survey of Causal Inference Applications at Netflix
2024	Video	Netflix	Graph Search	Reverse Searching Netflix’s Federated Graph
2024	Video	Netflix	A/B Testing	Sequential A/B Testing Keeps the World Streaming Netflix Part 1: Continuous Data
2024	Video	Netflix	A/B Testing	Sequential A/B Testing Keeps the World Streaming Netflix Part 2: Counting Processes
2023	Video	Netflix	Speech & Music	Detecting Speech and Music in Audio Content
2023	Video	Netflix	Video	Building In-Video Search
2023	Video	Netflix	Video	The Next Step in Personalization: Dynamic Sizzles
2023	Video	Netflix	ML Infra	Building a Media Understanding Platform for ML Innovations
2023	Video	Netflix	ML Infra	Scaling Media Machine Learning at Netflix
2023	Video	Netflix	Causal Inference	Causal Machine Learning for Creative Insights
2022	Video	Netflix	Fraud Detection	Machine Learning for Fraud Detection in Streaming Services
2022	Video	Netflix	Reinforcement Learning	Reinforcement Learning for Budget Constrained Recommendations
2021	Video	Netflix	A/B Testing	What is an A/B Test?
2021	Video	Netflix	A/B Testing	Interpreting A/B test results: false positives and statistical significance
2021	Video	Netflix	A/B Testing	Interpreting A/B test results: false negatives and power
2024	Music	Spotify	Annotations	How We Generated Millions of Content Annotations
2024	Music	Spotify	A/B Testing	Fixed-Power Designs: It’s Not IF You Peek; It’s WHAT You Peek at
2023	Music	Spotify	Causal Infernece	Experimenting with Machine Learning to Target In-App Messaging
2023	Music	Spotify	Clusterig	Recursive Embedding and Clustering
2022	Music	Spotify	Search	Introducing Natural Language Search for Podcast Episodes
2021	Music	Spotify	Personalization	How Spotify Uses ML to Create the Future of Personalization
2021	Music	Spotify	Personalization	The Rise (and Lessons Learned) of ML Models to Personalize Content on Home (Part I)
2021	Music	Spotify	Personalization	The Rise (and Lessons Learned) of ML Models to Personalize Content on Home (Part II)

5. Generic Topics

• [evidentlyai.com] AI quality and MLOps tutorials

• [recsys.acm.org] Tutorials on ACM RecSys Conferences

• [Eugene Yan] applied-ml repo on github

• [Updated] ApplyingML

• [Updated] What We’ve Learned From A Year of Building with LLMs

6. Resources

1. Designing Machine Learning Systems - Chip Huyen

2. Feature Engineering - O’Reilly

3. [deeplearning.ai] Machine Learning Yearning - Andrew Ng

4. [towardsdatascience.com] Being a Data Scientist does not make you a Software Engineer! - Towards Data Science

5. [developers.google.com] Machine Learning Crash Course - Google Developers

6. [tableau.com] 10 Great Machine Learning (ML) and Artificial Intelligence (AI) Blogs To Follow