Может ли плохая модель победить хорошую? Как Data Leakage обманывает метрики
Представим обычную ситуацию: мы обучили несколько моделей, сравнили их на тестовой выборке и получили довольно странный результат. Простейшее дерево…
Programming
⚑ Сообщить о проблемеПоследние новости Programming от Tech News
Все темы
AI
agents
ai
api
architecture
automation
backend
beginners
career
database
devchallenge
devops
discuss
javascript
llm
machinelearning
mcp
opensource
performance
productivity
programming
python
react
rust
security
showdev
testing
tutorial
typescript
webdev
Машинное обучение на PHP: Rubix ML 3.x – большое обновление экосистемы
В PHP-экосистеме не так часто происходят заметные изменения в области машинного обучения. Пока основное внимание разработчиков сосредоточено на Pyth…
Как обучить модель для upscale на ноутбуке с 4 ГБ и не умереть
Зачем нам свой улучшатор? Мы живем в эпоху доступности генеративных нейросетей, однако большинство качественных инструментов улучшения качества фото…
Presentation: Building GenAI Platform at DoorDash
Swaroop Chitlur and Sidd Kodwani share DoorDash’s journey building an internal GenAI platform. They discuss core architectural bets, transitioning f…
GKE Pod Snapshots Cut Model Load Times, and Move the Work to Snapshot Lifecycle Management
Google has published benchmarks for GKE Pod snapshots, reporting up to 89% lower startup latency and a 70B model loading in 37 seconds. The feature…
Presentation: Adaptive Recommenders in the Real World: Inference, Evals, and System Design
Mallika Rao explains that the true complexity of adaptive recommendation systems lies outside model architecture. She discusses how real-time feedba…
Apple Reference Image Signs Photos at the Sensor, Moving Provenance Trust Away from C2PA
Apple has published the design of Reference Image, an iPhone 18 Pro camera mode that signs pixel data at the sensor and develops it in Private Cloud…
Article: Beyond Relevance: A Governance-First Architecture for Enterprise Personalization
This article examines the limitations of conventional personalization systems and highlights the need for a governance-first architecture. It emphas…
GPT-6 Astra Is the First Model OpenAI Classifies as Critical for Cybersecurity
OpenAI has classified GPT-6 Astra at the Critical cybersecurity threshold under its Preparedness Framework, a first. In expert-led testing the model…
Presentation: From S3 to GPU in One Copy: Rethinking Data Loading for ML Training
Onur Satici explains how Vortex, an open-source columnar file format under the Linux Foundation, revolutionizes high-throughput data loading. He det…
Cohere’s Parse 5 Promises Efficient Multi-Modal Information Extraction From Complex Documents
Cohere has launched Parse 5, a multimodal foundation model designed to extract structured data from complex enterprise documents. The 2.3-billion-pa…
Swiggy Uses 350+ Features and Multi-Task MLP to Predict Customer Lifetime Value
Swiggy developed an in house predicted lifetime value model using more than 350 pre order features and a multi task MLP for Food and Instamart. Addi…
Presentation: Running AI at the Edge: Running Real Workloads Directly in the Browser
James Hall discusses the strategic and technical imperative of moving AI workloads from cloud providers to local edge devices. He shares practical a…
Presentation: SafeChat: Building AI-Powered Safety Systems at Scale in a Real-Time Marketplace
Bruna Pereira explains how DoorDash built a content-agnostic AI moderation platform. She covers replacing costly LLM-only pipelines with a hybrid pa…
Whatsapp Tests on Device ML for Scam Detection with Privacy Preserving Analytics
WhatsApp is testing Scam Alert in limited beta, using on device machine learning to detect potential scam messages from non contacts. Meta's archite…
Major Frontier Model Providers Adopt Watermarking Tech to Comply with EU Regulation
As of August 2, 2026, the EU AI Act Article 50 requires AI systems to mark synthetic outputs in a machine-detectable manner. Major vendors are imple…
Presentation: From Models to Agents: Building Context-Aware Consumer AI at Scale at DoorDash
Sudeep Das shares how DoorDash shifts from legacy one-shot predictions to an agentic recommendation platform. He discusses leveraging language-nativ…
Побеждаем OOM в PyTorch: как обучать гигантские графы на обычной видеокарте
При обучении графных нейросетей и гигантских эмбеддингов на миллионы узлов стандартный torch.optim.SparseAdam в PyTorch моментально забивает RAM и V…
Yelp Unifies ML Model Training with Training Orchestrator
Yelp has launched Training Orchestrator. This new internal framework replaces individual team Spark training scripts. Now, it uses a configuration-d…
Google Cloud Workbench Notebooks Extension Connects VS Code to Google Cloud's Jupyter Notebooks
The Google Cloud Workbench Notebooks extension for VS Code is a new tool that enables developers to connect their local IDE directly to managed Jupy…
Inside Target’s LLM-Based System for Semantic Matching in Marketing Forecast Pipelines
Target built a generative AI system to improve marketing campaign forecasting by retrieving and ranking similar historical campaigns. Using embeddin…
Article: The Technology Adoption Curve, Twenty Years On
Today, June 8th, InfoQ celebrates 20 years. This is not a comprehensive history, but a deliberately selective look at the technologies and practices…
Uber Improves Restaurant Recommendations Using Real-Time Signals and Listwise Ranking
Uber updates its Uber Eats Home Feed recommendation system using near real-time user sequence features and a Generative Recommender model. The syste…