Habr AI
AI news source. Articles are auto-selected and adapted by Hamidun News editors.
Latest publications
Target Encoding Without Data Leakage: LOO and K-Fold vs. the Illusion of Quality
Naive target encoding quietly leaks the target into the features and inflates metrics — a breakdown of exactly where leakage occurs and how to eliminate it with LOO and K-Fold.
MWS AI and SberAI Propose DRAGOn — a Dynamic Benchmark for Evaluating RAG Systems
Researchers from MWS AI, SberAI, and universities described DRAGOn — an approach to dynamic RAG evaluation where the corpus is regularly updated and questions and leaderboard are built automatically.
WebAsk launched an MCP server for surveys and found that AI reads more than it creates
WebAsk gave Claude and Cursor direct access to its survey builder via MCP, but the main use case unexpectedly turned out not to be questionnaire generation, but reading and analyzing hundreds of responses.
Habr AI explains how coding agents use memory, tools, and repository context
Habr AI's piece shows why the quality of a coding assistant depends not only on the model, but also on the harness, memory, tools, and access to the actual repository context.
Russian organizations showed six real AI use cases for project management
Russian organizations use AI in project management not universally, but in six clear scenarios: from knowledge bases and forecasts to automated meeting minutes and plan checks.
OpenClaw gets a Go reimplementation: one 35 MB binary instead of 800 MB of dependencies
The popular OpenClaw now has an independent Go version: it runs as a single 35 MB file, requires 3–5 times less memory, and simplifies deployment on low-end VPS and Raspberry Pi.
GitClear: AI speeds up code releases, but grows 'understanding debt' and hidden risks
GitClear analyzed 211 million lines of code and showed that AI speeds up releases, but at the same time increases churn, duplication, and code the team does not have time to understand.
ACE-Step 1.5 from ACE Studio outperforms Suno v5 and runs music generation locally
The open-source model ACE-Step 1.5 generates music locally with 4 GB of VRAM, produces a full track in seconds, and scores higher than Suno v5 on SongEval.
SpaceX, Blue Origin and Starcloud target orbital data centers for AI
SpaceX, Starcloud and Blue Origin want to move AI computing into orbit: near-continuous solar power generation could make space a new home for data centers.
Anthropic and Amazon show how AI at work erodes skills
New cases from software development and medicine show that constant reliance on generative AI speeds up individual tasks but gradually worsens understanding, debugging, and critical thinking.
TAPe raised classification accuracy to 77% and compared the results with YOLO on a small COCO dataset
The TAPe authors showed how object-boundary segmentation, abandoning the learning rate, and refining classification boosted performance on COCO, and why YOLO barely converges on a small dataset.
OTUS explained on Habr how AI agents for software development work: from tokens to tools
Habr published a detailed breakdown of the architecture of AI agents for software development: what building blocks they consist of, why the cost of each step is rising, and how tools with reasoning affect answer quality
A Habr author tested whether ChatGPT can reconstruct articles from short prompts
An experiment on Habr showed that ChatGPT can almost completely reconstruct an article from a short prompt if the text contains no unique facts gathered manually.
Telegram has become the main environment for AI agents: environment matters more than model quality
Telegram is becoming the main environment for AI agents: it already has users, communication, and context — everything an agent needs for real results.
LangChain in production: Habr AI explained why multi-agent systems are moving to plain Python
Habr AI published an analysis of a production multi-agent system without LangChain: the author shows why model switching, RAG, tool calling, and fallback require explicit logic rather than a 'magical' abstraction.
VK presented three student projects from EMS — from a multimodal benchmark to text analysis
VK presented three projects by EMS students: a multimodal benchmark for Russian, a lower-cost attention mechanism, and a system for analyzing emotions and states from text.
VTB opens Data Fusion 2026: robots, HiRAG, and a debate over applied AI
On the first day of VTB's Data Fusion 2026 at the Lomonosov cluster, the main themes were robots on site, HiRAG for VK Video, and the AI market's shift toward more practical use cases.
Surf’s Head of Frontend built a game with AI without animation experience — and nearly broke down
Andrey Makar-Uvarov of Surf tested a hypothesis: how much AI helps in an unfamiliar domain. He took on creating animation for an indie game — and got an honest lesson in the limits of vibe coding.
OpenClaw in production: Docker, Kubernetes, and fault tolerance under peak loads
How to scale OpenClaw for real-world load: containerization in Docker, a fault-tolerant Kubernetes cluster, and stateful storage to synchronize state between instances.
Habr AI showed how the virtual camera for Shorts and Reels with face tracking works
Habr AI broke down the virtual camera algorithm that turns horizontal video into a vertical format, keeps a face in frame, and smooths out jerks like a human camera operator.
Koda updated koda-cli 0.3.1: the assistant can now be integrated into a development environment
In koda-cli 0.3.1, Koda enhanced support for Agent Client Protocol so developers can launch and use the AI assistant directly inside the interface of a compatible development environment.
Habr AI explained why ChatGPT shouldn't write a business's technical specification for AI implementation
A ChatGPT-generated technical specification for AI implementation may look professional, but miss business context and data risks while blurring the criteria for choosing a contractor.
Habr AI and BotHub compiled 11 AI tools for text generation with a focus on Russian
In a new roundup, Habr AI listed 11 text-generation services — from BotHub and GigaChat to Copy AI — with an emphasis on Russian-language use cases, templates, and trial plans.
LangChain introduced Deep Agents — a layer on top of LangGraph for long-running and complex tasks
LangChain released Deep Agents, a Python library on top of LangGraph that adds planning, subagents, a file system, and memory, so developers do not have to assemble complex AI agents by hand.
Habr AI: how fear of AI-driven job loss is becoming a strategy for professionals to adapt
Habr has published a column on how fear of losing one’s profession to AI is giving way to practice: neural networks help with routine tasks, planning, and professional reorientation.
Sinimex presented an AI coach for enterprise sales with document-based verification
Sinimex described a prototype AI coach for enterprise sales managers: it conducts a negotiation dialogue, checks talking points against documents, and has already recorded 7 successful tests out of 8.
STMicroelectronics' STM32N6 demonstrated on-device speech recognition without the cloud at 0.2 W
An experiment on the STM32N6 with an integrated NPU showed that the microcontroller can recognize free-form speech locally, almost in real time, with power consumption of about 0.2 W.
DiffQuant optimizes the Sharpe ratio directly through a differentiable trading simulator
The open DiffQuant prototype closes the gap between ML proxy objectives and the real metric: the entire trading pipeline is assembled into a single graph, delivering Sharpe +1.73 and +1.15 on held-out data.
Codex Turned Into a Personal Assistant With Memory on Markdown, Git, and Telegram
Habr AI published a detailed breakdown of how Codex was turned into a personal assistant with long-term memory, roles, and automations without vector databases or complex RAG.
AvatarBox on HeyGen turns photos into talking videos right in Telegram in 2 minutes
Powered by HeyGen, AvatarBox turns a photo and text into a short video right in Telegram: you can choose a voice, frame format, emotions, and get a clip with auto-generated subtitles.