Habr AI
AI news source. Articles are auto-selected and adapted by Hamidun News editors.
Latest publications
RAGAS and RAG metrics: how to stop guessing and start measuring quality
The third part of our quality engineering series: how to automate the evaluation of RAG systems with RAGAS and monitor completeness, precision, and faithfulness of outputs instead of relying on manual review.
AI agent in one day: a local prototype without the cloud or developers
A practical guide for companies that want to quickly deploy an AI agent without the cloud, large budgets, or a team of developers: Ollama, n8n, and prompt engineering in one working day.
Rosenblatt's Perceptron: How the History of AI Began
The story of the first working neural network, the perceptron, created by Frank Rosenblatt in 1958, and its role in the beginning of the modern AI era.
How DocAI adapts medical education for each student
DocAI is a Russian AI-based project for medical education that automatically adapts content to each student's level and provides a personalized learning path.
AI image generators aren't creative: why and what to do
Illustrations for a text can be generated in a couple of minutes. But why does the result often look generic? Because neural networks choose statistically likely solutions, not creative ones.
How Creatures (1996) created the first true computer creatures
Steve Grand refused to write AI. Instead, he wrote genetics, emotion simulation, and neurons. The result became a cultural phenomenon.
Can LLMs find flaky tests from code? Study says no
A study found that LLMs are poor at identifying unreliable tests, even when dataset metrics suggest otherwise.
Computer vision platform that requires no GPUs or ML specialists
Developers unveiled a platform for training computer vision models on small datasets — a dozen images is enough to achieve 86% accuracy without GPUs or ML expertise.
Axera AX650N: Edge AI chip for local robotics instead of Jetson
Axera has unveiled the AX650N Edge AI chip for local neural network execution in robotics. An alternative to the costly NVIDIA Jetson, it runs YOLO, LLM and VLM directly on the device without the cloud.
LLMs write Rust with seven categories of hidden errors the compiler doesn’t catch
After six months of using Claude, GPT, and Cursor for Rust development, seven categories of errors were identified that the compiler misses but that compromise code safety or architecture.
Managing AI agents is now easier: the best open-source interfaces
User-friendly open-source tools with graphical interfaces have emerged for managing a fleet of AI agents. They help build task pipelines and distribute work among agents without programming.
How BERTopic with a local LLM helps Rostelecom analyze large text collections
An NLP engineer at Rostelecom shows a pipeline for clustering large volumes of text and generating interpretable topic names using BERTopic and a local LLM.
Several LLMs in one Telegram chat: how to combine Groq and Google AI
A developer created a Telegram bot where several LLM models work in a single chat and pretend to be corporate employees. Users switch between them on demand, using only free APIs.
Harness around an LLM delivers multi-fold gains: what changed after a year with Claude Code
An engineer who spent a year with Claude Code found that the harness around the model — system prompt, tools, context and skills — delivers a much larger quality gain than upgrading the model itself.
How Alpina Digital shielded AI from Russia's data law
88% of companies use AI, but only 1% have reached maturity. Alpina Digital found a solution: an architecture where requests go to the cloud while processing stays local. The result: 4.5x faster and $700,000 saved.
A neural network is a tool, not a teammate: how to work with AI properly
A developer often projects human work patterns onto a neural network — that is the main mistake. Habr explains why AI is not a second programmer and how to split tasks properly.
Bot4Web automates sales: how an AI chatbot works as a sales manager
Leo Lukin, a RAEC AI expert, automated lead generation, consultations, and service sales through the Bot4Web AI chatbot — now his website operates 24/7 without expanding the team.
The wave of AI automation has reached business. But who will manage it all?
Demand for AI automation is growing, hundreds of solutions are appearing on the market, but few explain to businesses how to manage deployed automation. This is the market’s key blind spot.
Inside an AI-native company: when development is a context conveyor belt
In AI-native companies, tools are just the facade. In reality, 80% of the work goes into passing information between people, documents, and decisions. How do you change that?
Dawkins talked with Claude: do LLMs have consciousness?
Biologist Richard Dawkins published an essay on the possible consciousness of language models, using Claude as an example. But people believe in AI consciousness based on empty dialogues, where words about the soul do no
Production AI agent architecture: Anthropic’s complete 13-block system
Anthropic published two real production prompts. We analyzed them and compiled a 13-block checklist: role, budget, OODA loop, division of responsibilities, resource ceilings.
Habr AI explained linear algebra for neural networks with practical examples and code
Habr AI released an introductory piece on linear algebra for those who want to understand neural networks: from the concepts of vectors and norms to operations, data vectorization, and hands-on Python practice.
Why the main AI revolution is not in code, but in language
Habr has published a piece arguing that AI's main breakthrough is not the quality of autogenerated code, but a new way of working: for the first time, a computer has begun to understand and accept human natural language.
Chinese AI models overtook U.S. ones in token consumption — OpenRouter data
Chinese models generated 4.69 trillion tokens versus 3.29 trillion for U.S. ones — for the second week in a row. OpenRouter is tracking a historic shift, bolstered by the mysterious Hunter Alpha.
How PMs can improve their technical skills with AI and build a real SaaS
A practical guide for product managers: how to use Vibe Coding and AI to build a SaaS product from scratch and start speaking the same language as developers.
OpenClaw gets a complete Russian-language guide to JSON5 config, CLI, and secure setup
A detailed guide to OpenClaw has been published on Habr: a single article brings together the JSON5 config structure, CLI commands, hot reload rules, channel management, and secure secret storage.
Russia’s Ministry of Digital Development, the EU and NIST set new rules for AI security and regulation
A major Security Vision review shows how Russia, the EU and the US are building rules for AI, which standards are already in effect, and why prompt injection, data leaks and Shadow AI are becoming key risks.
Alfa-Bank turned a tester’s idea into an AI stub generator for testing teams
A tester at Alfa-Bank built an API stub generator based on LLM, WireMock, and Spring AI, and the internal experiment grew into a product already being tested by the bank’s focus groups.
Yandex Practicum explained how data analysts can use AI without compromising quality
Yandex Practicum says AI is already speeding up the work of data analysts, but it does not replace knowledge of SQL, business metrics, or the ability to manually verify a model's answers.
Developer on Habr left an international company over its refusal to actively adopt AI
The author of a post on Habr said he left an international company after management refused to adopt AI: the business decided to keep building familiar web projects, while he chose LLMs, agents, and new products.