A-Curated-List-of-ML-System-Design-Case-Studies
A-Curated-List-of-ML-System-Design-Case-Studies 是一个专注于机器学习系统设计的实战案例库,汇集了来自 Netflix、Airbnb、Doordash 等 80 多家领先企业的 300 多个真实应用案例。它主要解决了开发者在构建 ML 系统时缺乏参考架构、难以获取一线生产环境经验的痛点,将分散在各公司技术博客和论文中的宝贵知识进行了系统化整理。
这份资源非常适合机器学习工程师、系统架构师以及相关领域的研究人员使用。无论是想深入了解推荐系统、搜索排序、计算机视觉还是自然语言处理的具体落地方案,用户都能在这里找到针对性的参考。其独特亮点在于内容的“真实性”与“深度”:每个案例均源自企业公开的技术复盘,详细拆解了目标用户、模型设计思路、评估指标以及部署架构等关键环节,而非泛泛而谈的理论介绍。通过按行业和具体应用场景分类的组织方式,用户可以快速定位到自己关心的领域,从他人的成功实践或踩坑经验中汲取灵感,从而更高效地设计出稳健的机器学习系统。
使用场景
某电商初创公司的算法团队正着手构建实时个性化推荐系统,急需参考成熟架构以避免重复造轮子。
没有 A-Curated-List-of-ML-System-Design-Case-Studies 时
- 盲目摸索架构:团队只能零散搜索博客,难以找到针对“冷启动”或“高并发排序”的具体工业级解决方案,导致技术选型犹豫不决。
- 忽视工程陷阱:缺乏对模型部署、数据漂移监控等生产环境细节的了解,极易在后期遭遇性能瓶颈或维护危机。
- 评估标准缺失:不清楚行业巨头如何定义推荐系统的成功指标(如点击率提升与多样性的平衡),导致内部评估体系片面。
- 沟通成本高昂:在向非技术背景的产品经理论证方案可行性时,缺乏权威的真实案例作为支撑,说服难度大。
使用 A-Curated-List-of-ML-System-Design-Case-Studies 后
- 精准对标方案:直接检索到 Netflix 和 Airbnb 在相似场景下的架构设计,快速确立了基于两塔模型的召回与排序流程。
- 规避已知风险:通过研读 DoorDash 等公司的实战复盘,提前规划了特征存储的一致性校验机制,避免了常见的线上故障。
- 建立科学指标:参考多家公司的评估框架,制定了兼顾短期转化与长期用户留存的多元化考核标准。
- 高效达成共识:利用库中详实的行业案例作为论据,直观展示技术路径的成熟度,迅速获得了管理层的资源支持。
A-Curated-List-of-ML-System-Design-Case-Studies 将分散的行业智慧转化为可执行的工程蓝图,显著降低了机器学习系统从设计到落地的试错成本。
运行环境要求

快速开始
机器学习系统设计案例库
简介
欢迎来到机器学习系统设计案例库!本仓库汇集了来自80多家领先公司的300余篇案例研究,展示了机器学习(ML)系统设计的实践应用与深入洞察。Netflix、Airbnb和Doordash等公司分享了他们的实践经验,为所有对如何利用机器学习改进产品和流程感兴趣的人士提供了宝贵资源。
🌐 精选资源:HorizonX.live
用HorizonX.live加速你的科研之旅——一站式满足你所有的科研需求。
- 🔍 搜索与探索:即时查找热门的机器学习论文、代码和数据集。
- 🗂️ 个性化文献库:保存、整理并标注你喜爱的研究资料。
- 📝 交互式论文摘要:每篇论文都配有简洁易懂的摘要。
🚀 即将推出的功能
我们正不断努力完善这一资源!以下是一些即将上线的精彩功能:
🌟 即将推出的功能
🟢 情境感知文献引擎
借助能够理解你研究背景的AI,加速你的文献综述工作。🤖 AI驱动的头脑风暴引擎
与你的智能助手协作,激发、塑造并提升突破性创意。🛠️ 低代码数据分析
无需编程即可将原始数据转化为可发表的图表与洞察。👥 实时协作
Google Docs与科研级Overleaf的结合,实现无缝团队合作。🔗 统一知识枢纽
一个智能工作空间,连接你的想法、洞见和科研工具。📚 自动化引用管理器
智能参考文献管理系统,轻松搞定引用。📝 智能格式化助手
让论文符合期刊要求变得轻而易举。🔍 出版前质量检查
作为你24/7的同行评审伙伴,确保稿件质量。💻 按需高性能计算访问
无需繁琐的配置即可获得强大的计算能力。
敬请期待这些更新,并随时提出你的建议或贡献内容!
立即体验: https://horizonx.live
功能特点
- 覆盖广泛行业:探索来自科技、金融、医疗健康等多个行业的案例研究。
- 多样化的机器学习应用:了解计算机视觉(CV)、自然语言处理(NLP)、推荐系统、搜索与排序、欺诈检测等众多机器学习应用场景。
- 具体产品功能:发现机器学习如何赋能各类面向用户的特性,例如语法错误修正和穿搭搭配生成等。
为何此资源极具价值
- 真实且深入:每篇案例均源自关于内部开发的机器学习系统的详尽博客、论文或文章,提供真实可靠的第一手见解。
- 实用性强:这些案例覆盖了实际生产环境中正在使用的机器学习系统,提供了切实可行且经过验证的示例。
- 聚焦且详尽:案例专注于特定的机器学习用例,清晰全面地介绍了目标用户、模型设计、评估标准及部署架构等内容。
使用指南
- 简要描述:通过简介快速找到与你兴趣相关的案例研究。
- 深入学习:仔细阅读详细的描述与实现细节,以更深入地理解机器学习系统设计。
- 分享与协作:如果你觉得这个数据库很有帮助,请将其推荐给他人,并通过建议新的案例来共同丰富该资源库。
祝您在这些案例研究中尽情探索,收获丰富的知识,进一步提升对机器学习系统设计的理解!
Real-world ML systems
| Index | Company | Industry | Description (< 5 words) | Title | Year |
|---|---|---|---|---|---|
| 1 | Stripe | Fintech and banking | Prevent fraudelent transactions | How we built it: Stripe Radar | 2023 |
| 2 | Walmart | E-commerce and retail | Recommend complementary items | Personalized ‘Complete the Look’ model | 2023 |
| 3 | Uber | Delivery and mobility | Forecast demand for airport rides | Demand and ETR Forecasting at Airports | 2023 |
| 4 | Social platforms | Prevent advertiser churn | An ML based approach to proactive advertiser churn prevention | 2023 | |
| 5 | Stitch Fix | E-commerce and retail | Generate ad headlines | A New Era of Creativity: Expert-in-the-loop Generative AI at Stitch Fix | 2023 |
| 6 | Swiggy | Delivery and mobility | Recommend items to order | Building a mind reader at Swiggy using Data Science | 2023 |
| 7 | Microsoft | Tech | Diagnose production incidents with LLM | Large-language models for automatic cloud incident management | 2023 |
| 8 | Foodpanda | Delivery and mobility | Optimize menu sorting order | Menu Ranking | 2023 |
| 9 | Zillow | E-commerce and retail | Estimate the house market value | Building the Neural Zestimate | 2023 |
| 10 | Airbnb | Travel,E-commerce and retail | Identify user interests | Prioritizing Home Attributes Based on Guest Interest | 2023 |
| 11 | GitHub | Tech | Generate code and code suggestions | Inside GitHub: Working with the LLMs behind GitHub Copilot | 2023 |
| 12 | DoorDash | Delivery and mobility | Optimize courier waiting time | Lifecycle of a Successful ML Product: Reducing Dasher Wait Times | 2023 |
| 13 | Social platforms | Select best payment gateway | Improving the customer’s experience via ML-driven payment routing | 2023 | |
| 14 | Wayfair | E-commerce and retail | Predict delivery times | Delivery-Date Prediction | 2023 |
| 15 | Social platforms | Detect viral spam | Viral spam content detection at LinkedIn | 2023 | |
| 16 | Lyft | Delivery and mobility | Recommend content in app | The Recommendation System at Lyft | 2023 |
| 17 | Honeycomb | Tech | Generate queries with natural language | All the Hard Stuff Nobody Talks About when Building Products with LLMs | 2023 |
| 18 | Zalando | E-commerce and retail | Forecast demand in fashion e-commerce | Deep Learning based Forecasting: a case study from the online fashion industry | 2023 |
| 19 | Etsy | E-commerce and retail | Recommend relevant marketplace items | How We Built a Multi-Task Canonical Ranker for Recommendations at Etsy | 2023 |
| 20 | Yelp | Social platforms | Organize e-commerce content using embeddings | Yelp Content As Embeddings | 2023 |
| 21 | Monzo | Fintech and banking | Select relevant marketing messages | Optimising marketing messages for Monzo users | 2023 |
| 22 | Monzo | Fintech and banking | Detect patterns in text data | Using topic modelling to understand customer saving goals | 2023 |
| 23 | Wayfair | E-commerce and retail | Predict new product’s sales potential | How Wayfair uses “Predicted Winners” Models to Accelerate Success for New Products | 2023 |
| 24 | Airbnb | Travel,E-commerce and retail | Personalized listing search | Learning To Rank Diversely | 2023 |
| 25 | Social platforms | Recommend interesting tweets | Twitter's Recommendation Algorithm | 2023 | |
| 26 | DoorDash | Delivery and mobility | Predict if a store is open | How DoorDash Upgraded a Heuristic with ML to Save Thousands of Canceled Orders | 2023 |
| 27 | Wayfair | E-commerce and retail | Identify business customers | Hamlet: Wayfair's ML Approach to Identifying Business Shopper | 2023 |
| 28 | Wayfair | E-commerce and retail | Detect fraud with embeddings | Introducing Melange: A Customer Journey Embedding System for Improving Fraud and Policy Abuse Detection | 2023 |
| 29 | Airbnb | Travel,E-commerce and retail | Improve travel search experience | Building Airbnb Categories with ML & Human in the Loop | 2023 |
| 30 | Spotify | Media and streaming | Automatically generate ad content | How We Automated Content Marketing to Acquire Users at Scale | 2023 |
| 31 | Instacart | E-commerce and retail | Predict availability of food items | How Instacart Modernized the Prediction of Real Time Availability for Hundreds of Millions of Items While Saving Costs | 2023 |
| 32 | Social platforms | Personalize the homepage feed | Enhancing homepage feed relevance by harnessing the power of large corpus sparse ID embeddings | 2023 | |
| 33 | Doordash | Delivery and mobility | Forecast order volumes and deliveries | How DoorDash Built an Ensemble Learning Model for Time Series Forecasting | 2023 |
| 34 | Expedia | Travel,E-commerce and retail | Forecast flight prices | Using Synthetic Search Data for Flights Price Forecasting | 2023 |
| 35 | Nextdoor | Social platforms | Generate engaging email subject lines | Let AI Entertain You: Increasing User Engagement with Generative AI and Rejection Sampling | 2023 |
| 36 | Criteo | Tech | Figure out users' preferences | Recommender systems need a user model | 2023 |
| 37 | Apple | Tech | Identify objects on images | Fast Class-Agnostic Salient Object Segmentation | 2023 |
| 38 | Zillow | E-commerce and retail | Identify and block unwanted callers | SpectroBrain: Detecting Phone Spam with Semi-Supervised Learning | 2023 |
| 39 | Algolia | Tech | Suggest relevant search queries | Feature Spotlight: Query Suggestions | 2023 |
| 40 | Netflix | Media and streaming | In-video search | Building In-Video Search | 2023 |
| 41 | Grab | Delivery and mobility,Banking and finance | Automatically tag sensitive data | LLM-powered data classification for data entities at scale | 2023 |
| 42 | Doordash | Delivery and mobility | Accurately forecast demand during holidays | How DoorDash Improves Holiday Predictions via Cascade ML Approach | 2023 |
| 43 | Netflix | Media and streaming | Personalize video clips | The Next Step in Personalization: Dynamic Sizzles | 2023 |
| 44 | BlaBlaCar | Delivery and mobility | Prevent phishing and payment fraud | How we used machine learning to fight fraud at BlaBlaCar — Part 1 | 2023 |
| 45 | Instacart | E-commerce and retail | Personalize user experience by recommending relevant products | Using Contextual Bandit models in large action spaces at Instacart | 2023 |
| 46 | Social platforms | Recommend similar visual content | Training Foundation Improvements for Closeup Recommendation Ranker | 2023 | |
| 47 | Spotify | Media and streaming | Recommend new complementary music | Spotify Track Neural Recommender System | 2023 |
| 48 | Meta | Social platforms | Generate code with LLM | Introducing Code Llama, a state-of-the-art large language model for coding | 2023 |
| 49 | Grammarly | Tech | Suggest gender-inclusive grammatical error corrections | Improving the Performance of NLP Systems on the Gender-Neutral “They” | 2023 |
| 50 | Netflix | Media and streaming | Detect speech and music in audio | Detecting Speech and Music in Audio Content | 2023 |
| 51 | Salesforce | Tech | Extract relevant information from a knowledge article | Resolve Cases Quickly with Interactive Einstein Search Answers | 2023 |
| 52 | Etsy | E-commerce and retail | Show relevant ads | Leveraging Real-Time User Actions to Personalize Etsy Ads | 2023 |
| 53 | GitHub | Tech | AI copilot for code generation | How to build an enterprise LLM application: Lessons from GitHub Copilot | 2023 |
| 54 | Uber | Delivery and mobility | Detect potential fraudulent entities | Risk Entity Watch – Using Anomaly Detection to Fight Fraud | 2023 |
| 55 | Expedia | Travel,E-commerce and retail | Predict Customer Lifetime Value (CLV) | Expedia Group’s Customer Lifetime Value Prediction Model | 2023 |
| 56 | Dailymotion | Media and streaming | Recommend diversified video content | Reinvent your recommender system using Vector Database and Opinion Mining | 2023 |
| 57 | Swiggy | Delivery and mobility | Predict food delivery time | Where is my order? — Part I | 2023 |
| 58 | Swiggy | Delivery and mobility | Сonversational and open-ended search | Swiggy’s Generative AI Journey: A Peek Into the Future | 2023 |
| 59 | New York Times | Media and streaming | Recommend recipes to readers | How The New York Times Cooking Team Makes Personalized Recipe Recommendations | 2023 |
| 60 | Expedia | Travel,E-commerce and retail | Suggest diverse travel recommendations | Generating Diverse Travel Recommendations | 2023 |
| 61 | Stitch Fix | E-commerce and retail | Personalize styling recommendations | Accelerating AI: Implementing Multi-GPU Distributed Training for Personalized Recommendations | 2023 |
| 62 | Doordash | Delivery and mobility | Areas for using Generative AI | DoorDash identifies Five big areas for using Generative AI | 2023 |
| 63 | Etsy | E-commerce and retail | Search by image | From Image Classification to Multitask Modeling: Building Etsy’s Search by Image Feature | 2023 |
| 64 | Spotify | Media and streaming | Generate audio podcast previews | Large-Scale Generation of ML Podcast Previews at Spotify with Google Dataflow | 2023 |
| 65 | Delivery Hero | Delivery and mobility | Better understand user behavior | Personalisation @ Delivery Hero: Understanding Customers | 2023 |
| 66 | Swiggy | Delivery and mobility | Predict food delivery time | Predicting Food Delivery Time at Cart | 2023 |
| 67 | Netflix | Media and streaming | Generate content recommendations for users | Lessons Learnt From Consolidating ML Models in a Large Scale Recommendation System | 2023 |
| 68 | Social platforms | Show relevant jobs in search | How LinkedIn Is Using Embeddings to Up Its Match Game for Job Seekers | 2023 | |
| 69 | Expedia | Travel,E-commerce and retail | Alert users about optimal deals | Increasing Travelers’ Engagement Through Price Alerts | 2023 |
| 70 | Walmart | E-commerce and retail | Resolve entities and detect relationships | Exploring an Entity Resolution Framework Across Various Use Cases | 2023 |
| 71 | Thoughtworks | Tech | AI copilot for product strategy | Building Boba AI | 2023 |
| 72 | Grab | Delivery and mobility,Banking and finance | Automatically detect new fraud types | Unsupervised graph anomaly detection - Catching new fraudulent behaviours | 2023 |
| 73 | Dropbox | Tech | Identify date formats in file names | Is this a date? Using ML to identify date formats in file names | 2023 |
| 74 | Grab | Delivery and mobility,Banking and finance | Сreate scalable lookalike audiences | Stepping up marketing for advertisers: Scalable lookalike audience | 2023 |
| 75 | Wayfair | E-commerce and retail | Send relevant communications to customers | Griffin: How Wayfair Leverages Reinforcement Learning to Send Customers Relevant Communications | 2023 |
| 76 | Whatnot | E-commerce and retail | Detect marketplace spam | How Whatnot Utilizes Generative AI to Enhance Trust and Safety | 2023 |
| 77 | Instacart | E-commerce and retail | Predict grocery item availability | How Instacart’s Item Availability Evolved Over the Pandemic | 2023 |
| 78 | Instacart | E-commerce and retail | Predict availability of food items | Instacart’s Item Availability Architecture: Solving for scale and consistency | 2023 |
| 79 | BlaBlaCar | Delivery and mobility | Prevent phishing and payment fraud | How we built our machine learning pipeline to fight fraud at BlaBlaCar — Part 2 | 2023 |
| 80 | Salesforce | Tech | Summarize Slack conversations | AI Summarist: Get Your Time Back on Slack, Boost Productivity & Focus, Personalize Information Consumption | 2023 |
| 81 | Meta | Social platforms | Show users relevant content at scale | Scaling the Instagram Explore recommendations system | 2023 |
| 82 | Delivery Hero | Delivery and mobility | Recommend restaurants for new customers | Personalisation @ Delivery Hero: Ranking restaurants for new users | 2023 |
| 83 | Swiggy | Delivery and mobility | Predict food delivery time | How ML Powers — When is my order coming? — Part II | 2023 |
| 84 | Salesforce | Tech | Recommend apps in the marketplace | On the Diversity and Explainability of Enterprise App Recommendation Systems | 2023 |
| 85 | Grab | Delivery and mobility,Banking and finance | Optimize promotional campaigns | Scaling marketing for merchants with targeted and intelligent promos | 2023 |
| 86 | GitHub | Tech | Automated code reviews and PR tagging | Generative AI-enabled compliance for software development | 2023 |
| 87 | Delivery Hero | Delivery and mobility | Recommend restaurants | Don’t Worry, We Got You: Personalised Model | 2023 |
| 88 | OLX | E-commerce and retail | Predict order delivery time | Machine Learning for Delivery Time Estimation | 2023 |
| 89 | Spotify | Media and streaming | Target in-app messaging | Experimenting with Machine Learning to Target In-App Messaging | 2023 |
| 90 | Nubank | Fintech and banking | Automatically route customer phone calls | Presenting Precog, Nubank’s Real Time Event AI | 2023 |
| 91 | Instacart | E-commerce and retail | Build an internal AI assistant | Scaling Productivity with Ava — Instacart’s Internal AI Assistant | 2023 |
| 92 | Meta | Social platforms | Translate and transcribe across speech and text | Bringing the world closer together with a foundational multimodal model for speech translation | 2023 |
| 93 | Vimeo | Media and streaming | Customer support AI assistant | From idea to reality: Elevating our customer support through generative AI | 2023 |
| 94 | Ebay | E-commerce and retail | Recommend relevant e-commerce items | Building a Deep Learning Based Retrieval System for Personalized Recommendations | 2022 |
| 95 | Mercado Libre | Delivery and mobility | Predict product dimensions for delivery | Predicting package dimensions based on a similarity model at Mercado Libre | 2022 |
| 96 | Doordash | Delivery and mobility | Recommend substitute items | Evolving DoorDash’s Substitution Recommendations Algorithm | 2022 |
| 97 | Social platforms | Personalize homepage contents | How Pinterest Leverages Realtime User Actions in Recommendation to Boost Homefeed Engagement Volume | 2022 | |
| 98 | Instacart | Delivery and mobility | Search food and grocery items | How Instacart Uses Embeddings to Improve Search Relevance | 2022 |
| 99 | Walmart | E-commerce and retail | Assist in e-commerce shopping | A Unified Multi-task Model for Supporting Multiple Virtual Assistants in Walmart | 2022 |
| 100 | Spotify | Media and streaming | Search for podcasts | Introducing Natural Language Search for Podcast Episodes | 2022 |
| 101 | Nextdoor | Social platforms | Predict harmful comments | Using predictive technology to foster constructive conversations | 2022 |
| 102 | Walmart | E-commerce and retail | Fill shopping cart via voice dialog | Voice Reorder Experience: add Multiple Product Items to your shopping cart | 2022 |
| 103 | Expedia | Travel,E-commerce and retail | Categorize customer feedback | Categorising Customer Feedback Using Unsupervised Learning | 2022 |
| 104 | Foodpanda | Delivery and mobility | Classify restaurants and cuisines | Classifying restaurant cuisines with subjective labels | 2022 |
| 105 | Ebay | Social platforms | Recommend products and content | Multi-Relevance Ranking Model for Similar Item Recommendation | 2022 |
| 106 | Gousto | Delivery and mobility | Predict subscription churn | Using Data Science to Retain Customers | 2022 |
| 107 | Tech | Generate summaries | Auto-generated Summaries in Google Docs | 2022 | |
| 108 | Yelp | Social platforms | Personalize recommendations | Beyond Matrix Factorization: Using hybrid features for user-business recommendations | 2022 |
| 109 | PayPal | Fintech and banking | Prioritize sales leads | Sales Pipeline Management with Machine Learning: A Lightweight Two-Layer Ensemble Classifier Framework | 2022 |
| 110 | Grubhub | Delivery and mobility | Forecast order volume | Forecasting Grubhub Order Volume At Scale | 2022 |
| 111 | Github | Tech | Detect vulnerabilities in code | Leveraging machine learning to find security vulnerabilities | 2022 |
| 112 | Uber | Delivery and mobility | Detect payment fraud | Project RADAR: Intelligent Early Fraud Detection System with Humans in the Loop | 2022 |
| 113 | Gojek | Delivery and mobility | Predict food delivery times | How We Estimate Food Debarkation Time With 'Tensoba' | 2022 |
| 114 | Uber | Delivery and mobility | Predict estimated time of arrival | DeepETA: How Uber Predicts Arrival Times Using Deep Learning | 2022 |
| 115 | Trivago | Travel,E-commerce and retail | Optimize accommodation ranking | Explore-exploit dilemma in Ranking model | 2022 |
| 116 | Gousto | Delivery and mobility | Recommend food items and recipes | Gousto R-series Vol 2: Tackling the Cold-Start Problem in Recipe Recommendation Engine | 2022 |
| 117 | Spotify | Media and streaming | Forecast user activity metrics | How We Built Infrastructure to Run User Forecasts at Spotify | 2022 |
| 118 | Tech | Summarize conversations | Conversation Summaries in Google Chat | 2022 | |
| 119 | Airbnb | Travel,E-commerce and retail | Improve travel search experience | Building Airbnb Categories with ML and Human-in-the-Loop | 2022 |
| 120 | Uber | Delivery and mobility | Send timely push notifications | How Uber Optimizes the Timing of Push Notifications using ML and Linear Programming | 2022 |
| 121 | Meta | Social platforms | Personalize daily digest notifications | Improving Instagram notification management with machine learning and causal inference | 2022 |
| 122 | Instacart | Delivery and mobility | Recommend relevant food items | Personalizing Recommendations for a Learning User | 2022 |
| 123 | Expedia | Travel,E-commerce and retail | Rank relevant travel deals | How to Optimise Rankings with Cascade Bandits | 2022 |
| 124 | Doordash | Delivery and mobility | Personalize recommendations on homepage | Homepage Recommendation with Exploitation and Exploration | 2022 |
| 125 | Social platforms | Improve post search functionality | Improving Post Search at LinkedIn | 2022 | |
| 126 | Artefact | Tech | Evaluate success of past promotions | Forecasting something that never happened: how we estimated past promotions profitability | 2022 |
| 127 | Doordash | Delivery and mobility | Find high-value merchants | Building the Model Behind DoorDash’s Expansive Merchant Selection | 2022 |
| 128 | Grammarly | Tech | Suggest text edits | Under the Hood of the Grammarly Editor, Part Two: How Suggestions Work | 2022 |
| 129 | Amazon | Media and streaming | Suggest music to listen to | The Amazon Music conversational recommender is hitting the right notes | 2022 |
| 130 | Snap | Social platforms | Rank relevant ads | Machine Learning for Snapchat Ad Ranking | 2022 |
| 131 | Instacart | E-commerce and retail | Autocomplete user searches in e-commerce | How Instacart Uses Machine Learning-Driven Autocomplete to Help People Fill Their Carts | 2022 |
| 132 | Zillow | E-commerce and retail | Select tags for product listings | Helping Home Shoppers Find a Home to Love Through Home Insights | 2022 |
| 133 | Netflix | Media and streaming | Detect account or content fraud | Machine Learning for Fraud Detection in Streaming Services | 2022 |
| 134 | Airbnb | Travel,E-commerce and retail | Improve customer support | How AI Text Generation Models Are Reshaping Customer Support at Airbnb | 2022 |
| 135 | Social platforms | Predict churn and upsell products | The journey to build an explainable AI-driven recommendation system | 2022 | |
| 136 | Autotrader | E-commerce and retail | Personalize automotive search results | Real-Time Personalisation of Search Results with Auto Trader's Customer Data Platform | 2022 |
| 137 | Peloton | Tech | Recommend fitness training videos | How We Built: An Early-Stage Machine Learning Model for Recommendations | 2022 |
| 138 | Walmart | E-commerce and retail | Categorize e-commerce products | Semantic Label Representation with an Application on Multimodal Product Categorization | 2022 |
| 139 | Doordash | Delivery and mobility | Search food and grocery items | 3 Changes to Expand DoorDash’s Product Search Beyond Delivery | 2022 |
| 140 | Faire | E-commerce and retail | Rank e-commerce items (feature store) | Real-time ranking at Faire part 2: the feature store | 2022 |
| 141 | New York Times | Media and streaming | Personalize paywall limits | How The New York Times Uses Machine Learning To Make Its Paywall Smarter | 2022 |
| 142 | Social platforms | Predict ad click-through rate | Challenges and practical lessons from building a deep-learning-based ads CTR prediction model | 2022 | |
| 143 | Zillow | E-commerce and retail | Identify customers that are likely to convert | Identifying High-Intent Buyers | 2022 |
| 144 | Netflix | Media and streaming | Recommend content to view | Reinforcement Learning for Budget Constrained Recommendations | 2022 |
| 145 | Walmart | E-commerce and retail | Forecast anomalies in refrigeration | Forecast Anomalies in Refrigeration with PySpark & Sensor-data | 2022 |
| 146 | Stitch Fix | E-commerce and retail | Recommend e-commerce items | Client Time Series Model: a Multi-Target Recommender System based on Temporally-Masked Encoders | 2022 |
| 147 | Gojek | Delivery and mobility | Predict estimated time of delivery | How We Estimate Food Debarkation Time With ‘Tensoba’ | 2022 |
| 148 | Zillow | E-commerce and retail | Extract text features | Incorporating Listing Descriptions into the Zestimate | 2022 |
| 149 | Etsy | E-commerce and retail | Rank marketplace search results | Deep Learning for Search Ranking at Etsy | 2022 |
| 150 | Walmart | E-commerce and retail | Curate e-commerce product recommendations | Scaling Product Recommendations using Basket Analysis- Part 1 | 2022 |
| 151 | Lyft | Delivery and mobility | Optimize trip price | Pricing at Lyft | 2022 |
| 152 | Grammarly | Tech | Correct grammatical errors | Innovating the Basics: Achieving Superior Precision and Recall in Grammatical Error Correction | 2022 |
| 153 | Social platforms | Recommend accounts to follow | Model-based candidate generation for account recommendations | 2022 | |
| 154 | Airbnb | Travel,E-commerce and retail | Improve customer travel experience | Intelligent Automation Platform: Empowering Conversational AI and Beyond at Airbnb | 2022 |
| 155 | Swiggy | Delivery and mobility | Flag incorrectly captured locations | Using deep learning to detect dissonance between address text and location | 2022 |
| 156 | Uber | Delivery and mobility | Verify documents | Uber’s Real-Time Document Check | 2022 |
| 157 | Wayfair | E-commerce and retail | Optimize email sending time and frequency | Nightingale: Scalable Daily Sales Email Sending Decision Model | 2022 |
| 158 | Didact AI | Fintech and banking | Predict stock prices | Didact AI: The anatomy of an ML-powered stock picking engine | 2022 |
| 159 | Wayfair | E-commerce and retail | Identify specific entities within a text | Wayfair’s New Approach to Aspect Based Sentiment Analysis Helps Customers Easily Find “Long Tail” Products | 2022 |
| 160 | Oda | Delivery and mobility | Predict driver's non-driving time | How we went from zero insight to predicting service time with a machine learning model — Part 2/2 | 2022 |
| 161 | Wayfair | E-commerce and retail | Predict intent in customer support messages | Building Wayfair’s First Virtual Assistant: Automating Customer Service by Text Based Intent Prediction | 2022 |
| 162 | Social platforms | Estimate the impact of product changes | Ocelot: Scaling observational causal inference at LinkedIn | 2022 | |
| 163 | Grab | Delivery and mobility,Banking and finance | Detect fraud with graph models | Graph for fraud detection | 2022 |
| 164 | Lyft | Delivery and mobility | Make causally valid forecasts | Causal Forecasting at Lyft (Part 1) | 2022 |
| 165 | Glassdoor | Social platforms | Recommend interesting posts to users | Personalized Fishbowl Recommendations with Learned Embeddings: Part 2 | 2022 |
| 166 | Netflix | Media and streaming | Improve video quality at scale | For your eyes only: improving Netflix video quality with neural networks | 2022 |
| 167 | Glassdoor | Social platforms | Recommend interesting posts to users | Personalized Fishbowl Recommendations with Learned Embeddings: Part 1 | 2022 |
| 168 | Dailymotion | Media and streaming | Recommend diversified video content | Optimizing video feed recommendations with diversity: Machine Learning first steps | 2022 |
| 169 | Siemens Healthineers | Tech | Optimize software testing | Using Machine Learning for Fast Test Feedback to Developers and Test Suite Optimization | 2022 |
| 170 | Lyft | Delivery and mobility | Make causally valid forecasts | Causal Forecasting at Lyft (Part 2) | 2022 |
| 171 | Social platforms | Deliver more relevant job recommendations | Improving job matching with machine-learned activity features | 2022 | |
| 172 | Cookidoo | E-commerce and retail | Personalize recipe recommendations | Building A Recipe Recommender System For the Thermomix on Cookidoo – Part 1 | 2022 |
| 173 | Social platforms | Improve ML model performance with multitask learning | Applying multitask learning to AI models at LinkedIn | 2022 | |
| 174 | Netflix | Media and streaming | Apply causality in experiments and marketing | A Survey of Causal Inference Applications at Netflix | 2022 |
| 175 | Social platforms | Recommend bids for advertizers | Advertiser Recommendation Systems at Pinterest | 2021 | |
| 176 | Grubhub | Delivery and mobility | Forecast volume order | “I See Tacos In Your Future”: Order Volume Forecasting at Grubhub | 2021 |
| 177 | Slack | Tech | Detect spam invites | Blocking Slack Invite Spam With Machine Learning | 2021 |
| 178 | Faire | E-commerce and retail | Search and navigate marketplace items | Building Faire’s new marketplace ranking infrastructure | 2021 |
| 179 | Doordash | E-commerce and retail | Predict delivery supply and demand | Managing Supply and Demand Balance Through Machine Learning | 2021 |
| 180 | OLX | E-commerce and retail | Recommend e-commerce items | Item2Vec: Neural Item Embeddings to enhance recommendations | 2021 |
| 181 | Dropbox | Tech | Search by image content | How image search works at Dropbox | 2021 |
| 182 | Scribd | Media and streaming | Extract metadata from documents | Information Extraction at Scribd | 2021 |
| 183 | Microsoft | Tech | Rank customer support cases | ML and customer support (Part 1): Using Machine Learning to enable world-class customer support | 2021 |
| 184 | Stitch Fix | E-commerce and retail | Recommend e-commerce inventory | Algorithm-Assisted Inventory Curation | 2021 |
| 185 | Social platforms | Forecast resource usage and cost | Forecasting SQL query resource usage with machine learning | 2021 | |
| 186 | Tech | Suggest past photos to look at | A snapshot of AI-powered reminiscing in Google Photos | 2021 | |
| 187 | Uber | Delivery and mobility | Identify cash intermediaries | Applying Machine Learning in Internal Audit with Sparsely Labeled Data | 2021 |
| 188 | Microsoft | Tech | Cluster customer support issues by similarity | ML and customer support (Part 2): Leveraging topic modeling to identify the top investment areas in support cases | 2021 |
| 189 | Gousto | Delivery and mobility | Recommend food items and recipes | Gousto R-series vol 1: Three tales of the Rouxcommender family | 2021 |
| 190 | Apple | Tech | Recognize people in photos | Recognizing People in Photos Through Private On-Device Machine Learning | 2021 |
| 191 | Social platforms | Find lookalike users for ad targeting | The machine learning behind delivering relevant ads | 2021 | |
| 192 | Social platforms | Detect spam users | Fighting Spam using Clustering and Automated Rule Creation | 2021 | |
| 193 | PayPal | Fintech and banking | Detect payment fraud | Deploying Large-scale Fraud Detection Machine Learning Models at PayPal | 2021 |
| 194 | Datto | Tech | Predict hard drive failures | Predicting Hard Drive Failure with Machine Learning | 2021 |
| 195 | Bumble | Social platforms | Detect rude messages | Multilingual message content moderation at scale (part 2) | 2021 |
| 196 | Nextdoor | Social platforms | Send relevant and timely updates | Nextdoor Notifications: How we use ML to keep neighbors informed | 2021 |
| 197 | Dropbox | Tech | Identify best time for renewal charge | Optimizing payments with machine learning | 2021 |
| 198 | Swiggy | Delivery and mobility | Rank restaurants in search | Learning To Rank Restaurants | 2021 |
| 199 | Brex | Fintech and banking | Classify bank transactions | How We Built a (Mostly) Automated System to Solve Credit Card Merchant Classification | 2021 |
| 200 | Grammarly | Tech | Capture what readers pay attention to | ATTN: How Grammarly’s NLP/ML Team Figured Out Where Readers Focus in an Email | 2021 |
| 201 | Doordash | Delivery and mobility | Extract information from images | How DoorDash Quickly Spins Up Multiple Image Recognition Use Cases | 2021 |
| 202 | Apple | Tech | Identify best user experience | Interpretable Adaptive Optimization | 2021 |
| 203 | Airbnb | Travel,E-commerce and retail | Data privacy and security | Automating Data Protection at Scale, Part 2 | 2021 |
| 204 | Capital One | Fintech and banking | Identify suspicious account activity | How Machine Learning Can Help Fight Money Laundering | 2021 |
| 205 | Wayfair | E-commerce and retail | Assign color names to products | From RGB to Descriptive Color Names: Wayfair's in-house color algorithms to improve customer shopping experience. | 2021 |
| 206 | Capital One | Fintech and banking | Automate incident management | Automated detection, diagnosis & remediation of app failure | 2021 |
| 207 | Social platforms | Detect policy-violating comments | How Pinterest powers a healthy comment ecosystem with machine learning | 2021 | |
| 208 | Spotify | Media and streaming | Personalize homepage content (podcasts, playlist, music) | The Rise (and Lessons Learned) of ML Models to Personalize Content on Home (Part I) | 2021 |
| 209 | Stitch Fix | E-commerce and retail | Recommend looks | Stitching together spaces for query-based recommendations | 2021 |
| 210 | Ocado | E-commerce and retail | Forecast e-commerce grocery demand | Finding the sweet spot | 2021 |
| 211 | Walmart | E-commerce and retail | Categorize e-commerce products | Deep Learning: Product Categorization and Shelving | 2021 |
| 212 | Walmart | E-commerce and retail | Recommend learning content | Mozrt, a Deep Learning Recommendation System Empowering Walmart Store Associates with a Personalized Learning Experience | 2021 |
| 213 | Walmart | E-commerce and retail | Identify refrigeration defrost | Predicting Defrost in Refrigeration Cases at Walmart using Fourier Transform | 2021 |
| 214 | New York Times | Media and streaming | Recommend content to read | Machine Learning and Reader Input Help Us Recommend Articles | 2021 |
| 215 | Mercado Libre | E-commerce and retail | Forecast demand for e-commerce items | Marketplace Forecasting: Sales or Demand? Why not both? Let’s find out! | 2021 |
| 216 | Swiggy | Delivery and mobility | Rank food dishes in search | Using Deep Learning for Ranking in Dish Search | 2021 |
| 217 | PayPal | Fintech and banking | Recommend financial products | Cross-Selling Optimization Using Deep Learning | 2021 |
| 218 | Wayfair | E-commerce and retail | Automate ads placement and bidding | Evolution of Ads Bidding at Wayfair | 2021 |
| 219 | Capital One | Fintech and banking | Improve cardholder experience | Improving Virtual Card Numbers with Edge Machine Learning | 2021 |
| 220 | Shopify | E-commerce and retail | Categorize e-commerce products | Using Rich Image and Text Data to Categorize Products at Scale | 2021 |
| 221 | Scribd | Media and streaming | Recommend content to read | Embedding-based Retrieval at Scribd | 2021 |
| 222 | Swiggy | Delivery and mobility | Detect fraud in online food delivery | DeFraudNet: An End-to-End Weak Supervision Framework to Detect Fraud in Online Food Delivery | 2021 |
| 223 | Amazon | E-commerce and retail | Predict coordinates of delivery location | Using learning-to-rank to precisely locate where to deliver packages | 2021 |
| 224 | PayPal | Fintech and banking | Predict declined transactions | Using Machine Learning to Improve Payment Authorization Rate | 2021 |
| 225 | Stripe | Fintech and banking | Detect fraud in online payments | A primer on machine learning for fraud detection | 2021 |
| 226 | Slack | Tech | Predict Slack connect invites | Email Classification | 2021 |
| 227 | Wayfair | E-commerce and retail | Recommend furniture items | MARS: Transformer Networks for Sequential Recommendation | 2021 |
| 228 | Grammarly | Tech | Detect grammatical errors | Grammatical Error Correction: Tag, Not Rewrite | 2021 |
| 229 | Nordstrom | E-commerce and retail | Generate outfit combinations | AI-Created Outfits | 2021 |
| 230 | Doordash | Delivery and mobility | Deliver orders on time | Using ML and Optimization to Solve DoorDash’s Dispatch Problem | 2021 |
| 231 | Zillow | E-commerce and retail | Recommend similar homes | Improving Recommendation Quality by Tapping into Listing Text | 2021 |
| 232 | Lifen | Tech | Recognize PDF layout | Fast graph-based layout detection | 2021 |
| 233 | PayPal | Fintech and banking | Prevent repeated payment fraud | How PayPal Uses Real-time Graph Database and Graph Analysis to Fight Fraud | 2021 |
| 234 | Bumble | Social platforms | Detect rude messages | Multilingual message content moderation at scale (part 1) | 2021 |
| 235 | Spotify | Media and streaming | Personalize homepage content (podcasts, playlist, music) | The Rise (and Lessons Learned) of ML Models to Personalize Content on Home (Part II) | 2021 |
| 236 | Swiggy | Delivery and mobility | Estimate travel distance | Learning to Predict Two-Wheeler Travel Distance | 2021 |
| 237 | Expedia | Travel,E-commerce and retail | Personalize travel search results | Personalized Ranking Model for Lodging | 2021 |
| 238 | Scribd | Media and streaming | Classify documents | Categorizing user-uploaded documents | 2021 |
| 239 | Meta | Social platforms | Personalize the newsfeed content | How machine learning powers Facebook’s News Feed ranking | 2021 |
| 240 | Tech | Correct grammatical errors | Grammar Correction as You Type, on Pixel 6 | 2021 | |
| 241 | Nubank | Fintech and banking | Predict conversions and attract new customers | Beyond prediction machines | 2021 |
| 242 | Grammarly | Tech | Correct grammatical errors | Adversarial Grammatical Error Correction | 2021 |
| 243 | Scribd | Media and streaming | Classify user-uploaded documents | Identifying Document Types at Scribd | 2021 |
| 244 | Oda | Delivery and mobility | Predict driver's non-driving time | How we went from zero insight to predicting service time with a machine learning model — Part 1 | 2021 |
| 245 | Mercado Libre | E-commerce and retail | Predict customer engagement and LTV | Causal Inference — Estimating Long-term Engagement | 2021 |
| 246 | Dailymotion | Media and streaming | Target contextual advertising | How Deep Learning can boost Contextual Advertising Capabilities | 2021 |
| 247 | Wayfair | E-commerce and retail | Optimize digital ads | Building Scalable and Performant Marketing ML Systems at Wayfair | 2021 |
| 248 | Wayfair | E-commerce and retail | Show relevant content to new customers | Share of Voice Optimization Engine | 2021 |
| 249 | Wayfair | E-commerce and retail | Optimize paid media marketing | Contextual Bandit for Marketing Treatment Optimization | 2021 |
| 250 | Microsoft | Tech | Classify cloud workload types | How we used ML — and heuristic data labeling — to help customers with their cloud migration | 2021 |
| 251 | Github | Tech | Help users find contribution opportunities | How we built the good first issues feature | 2020 |
| 252 | Social platforms | Serve personalized learning recommendations | A closer look at the AI behind course recommendations on LinkedIn Learning, Part 1 | 2020 | |
| 253 | Bumble | Social platforms | Derive information from images | Image detection as a service | 2020 |
| 254 | Gojek | Delivery and mobility | Generate names for pickup points | How Gojek Uses NLP to Name Pickup Locations at Scale | 2020 |
| 255 | Mozilla | Tech | Predict the outcome of software tests | Testing Firefox more efficiently with machine learning | 2020 |
| 256 | Adyen | Fintech and banking | Predict probability of transaction success | Optimizing payment conversion rates with contextual multi-armed bandits | 2020 |
| 257 | Wayfair | E-commerce and retail | Detect payment fraud | Explainable Fraud Detection | 2020 |
| 258 | Lyft | Delivery and mobility | Provide location suggestions | How Lyft predicts a rider’s destination for better in-app experience | 2020 |
| 259 | Zillow | E-commerce and retail | Generate floor plans from photos | Zillow Floor Plan: Training Models to Detect Windows, Doors and Openings in Panoramas | 2020 |
| 260 | Social platforms | Serve personalized learning recommendations | A closer look at the AI behind course recommendations on LinkedIn Learning, Part 2 | 2020 | |
| 261 | Doordash | Delivery and mobility | Optimize marketing spending | Optimizing DoorDash’s Marketing Spend with Machine Learning | 2020 |
| 262 | Etsy | E-commerce and retail | Personalize e-commerce search | Bringing Personalized Search to Etsy | 2020 |
| 263 | Airbnb | Travel,E-commerce and retail | Rank travel search results | Improving Deep Learning for Ranking Stays at Airbnb | 2020 |
| 264 | Wayfair | E-commerce and retail | Improve search experience for new customers | Bayesian Product Ranking at Wayfair | 2020 |
| 265 | Social platforms | Predict value of ad requests | Using machine learning to predict the value of ad requests | 2020 | |
| 266 | Zynga | Gaming | Personalize push notification timing | Deep Reinforcement Learning in Production Part 2: Personalizing User Notifications | 2020 |
| 267 | Zillow | E-commerce and retail | Rank homes to buy | Guided Search — Personalized Search Refinements to Help Customers Find their Dream Home | 2020 |
| 268 | Picnic | Delivery and mobility | Predict delivery drop times | Optimal drop times using machine learning | 2020 |
| 269 | Shopify | E-commerce and retail | Categorize e-commerce products | Categorizing Products at Scale | 2020 |
| 270 | Gojek | Delivery and mobility | Target cross-sell to existing users | How We Built a Matchmaking Algorithm to Cross-Sell Products | 2020 |
| 271 | PayPal | Fintech and banking | Detect payment fraud | Multi-Domain Fraud Detection While Reducing Good User Declines | 2020 |
| 272 | OLX | E-commerce and retail | Detect stolen photos | Fighting fraud with Triplet Loss | 2020 |
| 273 | Stripe | Fintech and banking | Detect fraud in online payments | Similarity clustering to catch fraud rings | 2020 |
| 274 | Doordash | Delivery and mobility | Search for restaurants and dishes | Things Not Strings: Understanding Search Intent with Better Recall | 2020 |
| 275 | Spotify | Media and streaming | Recommend shortcuts for homepage | Reach for the Top: How Spotify Built Shortcuts in Just Six Months | 2020 |
| 276 | Wayfair | E-commerce and retail | Recommend complementary products | The Visual Complements Model (ViCs): Complementary Product Recommendations From Visual Cues | 2020 |
| 277 | Dailymotion | Media and streaming | Automatically categorize videos | How we used Cross-Lingual Transfer Learning to categorize our content | 2020 |
| 278 | Duolingo | Tech | Teaching foreign languages | How Duolingo uses AI in every part of its app | 2020 |
| 279 | Firefox | Tech | Automatically assign new untriaged bugs | Teaching machines to triage Firefox bugs | 2019 |
| 280 | Dropbox | Tech | Predict files users search for | Using machine learning to predict what file you need next | 2019 |
| 281 | Zoominfo | Tech | Predict data accuracy | Using Machine Learning to Determine Contact Accuracy Scores | 2019 |
| 282 | Airbnb | Travel,E-commerce and retail | Recommend marketplace items | Machine Learning-Powered Search Ranking of Airbnb Experiences | 2019 |
| 283 | Lyft | Delivery and mobility | Predict location of traffic control elements | Detecting Stop Signs and Traffic Signals: Deep Learning at Lyft Mapping | 2019 |
| 284 | Gojek | Delivery and mobility | Personalize search results | The Secret Sauce Behind Search Personalisation | 2019 |
| 285 | Instacart | Delivery and mobility | Spot lost demand | Modeling the unseen | 2019 |
| 286 | Apple | Tech | Identify text language | Language Identification from Very Short Strings | 2019 |
| 287 | Stitch Fix | E-commerce and retail | Extract information from customer notes | Give Me Jeans not Shoes: How BERT Helps Us Deliver What Clients Want | 2019 |
| 288 | Lyft | Delivery and mobility | Detect errors in maps | How Lyft Creates Hyper-Accurate Maps from Open-Source Maps and Real-Time Data | 2019 |
| 289 | King | Gaming | Automate playtesting pipeline | Human-Like Playtesting with Deep Learning | 2019 |
| 290 | Gojek | Delivery and mobility | Analyse the relevance of search results | Is This What You Were Looking For? | 2019 |
| 291 | Lyft | Delivery and mobility | Build a marketing automation platform | Building Lyft’s Marketing Automation Platform | 2019 |
| 292 | Wayfair | E-commerce and retail | Model uplift | Modeling Uplift Directly: Uplift Decision Tree with KL Divergence and Euclidean Distance as Splitting Criteria | 2019 |
| 293 | Gojek | Delivery and mobility | Accurately forecast demand | Under the Hood of Gojek’s Automated Forecasting Tool | 2019 |
| 294 | Lyft | Delivery and mobility | Predict rides and driver hours | Making cohort-based long-term forecasts at Lyft | 2019 |
| 295 | Lyft | Delivery and mobility | Predict fraudulent activity | Fingerprinting fraudulent behavior | 2018 |
| 296 | Netflix | Media and streaming | Improve streaming quality | Using Machine Learning to Improve Streaming Quality at Netflix | 2018 |
| 297 | Lyft | Delivery and mobility | Identify user fraud | From shallow to deep learning in fraud | 2018 |
| 298 | Instacart | E-commerce and retail | Predict grocery item availability | Predicting the real-time availability of 200 million grocery items | 2018 |
| 299 | Lyft | Delivery and mobility | Personalize marketing offers | Empowering personalized marketing with machine learning | 2018 |
| 300 | Instacart | E-commerce and retail | Optimize food delivery logistics | Space, Time and Groceries | 2017 |
| 301 | Airbnb | Travel,E-commerce and retail | Predict Value of Homes | Using Machine Learning to Predict Value of Homes On Airbnb | 2017 |
| 302 | Netflix | Media and streaming | Improve Streamning Quality | Using Machine Learning to Improve Streaming Quality at Netflix | 2018 |
| 303 | Booking.com | Travel,E-commerce and retail | 150 Successful Machine Learning Models | 150 Successful Machine Learning Models: 6 Lessons Learned at Booking.com | 2019 |
| 304 | Chicisimo | Fashion and retail | Grow User base using vertical ML approch | How we grew from 0 to 4 million women on our fashion app, with a vertical machine learning approach | 2019 |
| 305 | Airbnb | Travel,E-commerce and retail | ML Powered search ranking | Machine Learning-Powered Search Ranking of Airbnb Experiences | 2019 |
| 306 | Lyft | Delivery and mobility | Shallow to deep learning in fraud | From shallow to deep learning in fraud | 2018 |
| 307 | Uber | Delivery and mobility | 100+ Petabytes with Minute Latency | Uber's Big Data Platform: 100+ Petabytes with Minute Latency | 2018 |
| 308 | Dropbox | Tech | Modern OCR with CV and DL | Creating a Modern OCR Pipeline Using Computer Vision and Deep Learning | 2017 |
| 309 | Uber | Tech | Scaling ML with Michelangelo | Scaling Machine Learning at Uber with Michelangelo | 2019 |
欲了解更多信息,请访问 Evidently AI - ML 系统设计 和 ML 系统设计
星标历史
相似工具推荐
stable-diffusion-webui
stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面,旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点,将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。 无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师,还是想要深入探索模型潜力的开发者与研究人员,都能从中获益。其核心亮点在于极高的功能丰富度:不仅支持文生图、图生图、局部重绘(Inpainting)和外绘(Outpainting)等基础模式,还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外,它内置了 GFPGAN 和 CodeFormer 等人脸修复工具,支持多种神经网络放大算法,并允许用户通过插件系统无限扩展能力。即使是显存有限的设备,stable-diffusion-webui 也提供了相应的优化选项,让高质量的 AI 艺术创作变得触手可及。
everything-claude-code
everything-claude-code 是一套专为 AI 编程助手(如 Claude Code、Codex、Cursor 等)打造的高性能优化系统。它不仅仅是一组配置文件,而是一个经过长期实战打磨的完整框架,旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。 通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能,everything-claude-code 能显著提升 AI 在复杂任务中的表现,帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略,使得模型响应更快、成本更低,同时有效防御潜在的攻击向量。 这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库,还是需要 AI 协助进行安全审计与自动化测试,everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目,它融合了多语言支持与丰富的实战钩子(hooks),让 AI 真正成长为懂上
ComfyUI
ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎,专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式,采用直观的节点式流程图界面,让用户通过连接不同的功能模块即可构建个性化的生成管线。 这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景,也能自由组合模型、调整参数并实时预览效果,轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性,不仅支持 Windows、macOS 和 Linux 全平台,还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构,并率先支持 SDXL、Flux、SD3 等前沿模型。 无论是希望深入探索算法潜力的研究人员和开发者,还是追求极致创作自由度的设计师与资深 AI 绘画爱好者,ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能,使其成为当前最灵活、生态最丰富的开源扩散模型工具之一,帮助用户将创意高效转化为现实。
NextChat
NextChat 是一款轻量且极速的 AI 助手,旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性,以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发,NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。 这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言,它也提供了便捷的自托管方案,支持一键部署到 Vercel 或 Zeabur 等平台。 NextChat 的核心亮点在于其广泛的模型兼容性,原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型,让用户在一个界面即可自由切换不同 AI 能力。此外,它还率先支持 MCP(Model Context Protocol)协议,增强了上下文处理能力。针对企业用户,NextChat 提供专业版解决方案,具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能,满足公司对数据隐私和个性化管理的高标准要求。
ML-For-Beginners
ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程,旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周,包含 26 节精炼课程和 52 道配套测验,内容涵盖从基础概念到实际应用的完整流程,有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。 无论是希望转型的开发者、需要补充算法背景的研究人员,还是对人工智能充满好奇的普通爱好者,都能从中受益。课程不仅提供了清晰的理论讲解,还强调动手实践,让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持,通过自动化机制提供了包括简体中文在内的 50 多种语言版本,极大地降低了全球不同背景用户的学习门槛。此外,项目采用开源协作模式,社区活跃且内容持续更新,确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路,ML-For-Beginners 将是理想的起点。
ragflow
RAGFlow 是一款领先的开源检索增强生成(RAG)引擎,旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体(Agent)能力相结合,不仅支持从各类文档中高效提取知识,还能让模型基于这些知识进行逻辑推理和任务执行。 在大模型应用中,幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构(如表格、图表及混合排版),显著提升了信息检索的准确度,从而有效减少模型“胡编乱造”的现象,确保回答既有据可依又具备时效性。其内置的智能体机制更进一步,使系统不仅能回答问题,还能自主规划步骤解决复杂问题。 这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统,还是致力于探索大模型在垂直领域落地的创新者,都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口,既降低了非算法背景用户的上手门槛,也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目,它正成为连接通用大模型与行业专有知识之间的重要桥梁。
