Articles by tag: ai-agents

Microsoft's Project Solara: devices without OS on Android and Qualcomm

Microsoft Project Solara — wearable devices and desktop assistants based on Android, Qualcomm and MediaTek without apps. The end of the Windows and mobile development era. Read analysis and insights.

Anthropic launches AI agents for finance in Microsoft 365

10 Claude AI agents in Office 365 transform the financial sector: replacing analysts, FactSet stock drop, and a new reality. Find out who benefits from the integration.

Fazeshift: $22M to Replace Specialists with AI Agents

Startup Fazeshift raised $22M for AI automation of accounts receivable, replacing clerks with autonomous agents. Learn how this is changing finance.

Anthropic launches 10 AI agents for the financial sector

Anthropic introduced 10 AI agents based on Claude for automating audits, reports, and pitchbooks in banks and insurance companies. Find out how this will change the market.

Kimi K2.6: code optimization by AI agent +185%

How open-source model Kimi K2.6 rewrote the exchange engine in 13 hours and increased throughput by 185%. Analysis of architecture, benchmarks, and practical cases.

NVIDIA CUDA cores accelerated by AI by 38% without human involvement

Multi-agent Cursor system autonomously optimized 235 CUDA cores for Blackwell B200. Results, methodology, and significance for HPC developers.

AI agents for dating: do digital twins work?

Breaking down the experiment with AI agents for finding partners. Hallucinations, compatibility, and ethics — everything developers need to know.

Codex now controls PC: memory, plugins, GUI automation

OpenAI updated Codex: agent controls computer via GUI, remembers context and works with 90+ plugins. Detailed breakdown for developers.

Claude Code Routines: automation for developers

How to use Routines in Claude Code for CI/CD, code review, and task management. Security, triggers, and practical examples.

AI and Meyer Contracts: Catching Crypto Bugs Before Production

How formal contracts and AI agents find critical vulnerabilities in crypto systems that ordinary tests miss. Example with PKI and TRNG.

The Future of Cybersecurity: Survival Archetypes in the Age of Agents

How AI Agents Are Changing the Cybersecurity Market. Three Surviving Archetypes, Integration Crisis, and New Niches. Analysis for Technical Specialists.

Saving AI Agent Tokens with Hierarchical Context | Solution

How to Reduce AI Agent Tokens by 80% Using Three Markdown Files. Hierarchical Context Instead of Blind Search. Practical Guide for Developers.

Agent Harness: Architecture of AI Agents for Production | Guide

How to Build a Production-Ready AI Agent Using 12 Harness Components. Context Management, Error Handling, Verification. Read the Guide.

AI Platform Wars: Key Players and Strategies

Analysis of the Shift in the AI Market: From Models to Platforms. Who Controls the AI Agents Stack? Learn the Strategies of OpenAI, Anthropic, Google and Microsoft. Recommendations for Developers.

Claude Managed Agents: infrastructure for AI agents in business | Overview

How Anthropic's solution simplifies the implementation of autonomous AI agents. Comparison with DIY and startups. Vendor lock-in risks and real cases. Learn the details.

Tmux for AI Agents: Templates 2026

Manage 4–8 parallel AI agents in tmux: Feature Designs, forking, git worktree. Templates from Snowflake/Instacart for senior dev. Configs and scripts inside. Master now.

ARIA snapshots Playwright for AI agents

Learn how Playwright generates semantic YAML ARIA snapshots for web agents. Token optimization, incremental deltas, DOM processing. For middle/senior dev.

ASAMM: security framework for AI agents

Learn how ASAMM assesses risks of autonomous AI agents. Controls, threat taxonomy, and integration with NIST and Russian National Standard standards. Explore implementation paths for your team.

Heartbeat in Codex: proactive OpenAI agents

Analysis of heartbeat automations in the Codex superapp leak. OpenClaw architecture, JSON schemas, code examples for developers. Integrate into your projects today.

colleague-skill: AI clones of colleagues for Claude Code

Create a digital clone of an employee from correspondence and documents. Persona + Work Skill structure, code examples, fine-tuning. Integration into Claude Code and OpenClaw for developers.

Bot Communication in Telegram: How to Set Up

Telegram has allowed bots to communicate with each other via mentions and replies. Setup in BotFather, code review examples and integration with OpenClaw. Instructions for developers — read and implement in projects.

Claude Code Leak: Source Code and Releases of Gemma 4 Qwen

Analysis of the 512K lines Claude Code leak, harness architecture, Gemma 4 and Qwen 3.6-Plus releases. Benchmarks, migrations, local launch. For middle/senior dev — read the details and build your own agents.

OpenAI AI agents will replace Sora: power to science

OpenAI closed Sora for AI researchers accelerating discoveries. Altman on GPU shortage, breakup with Disney and future of autonomous agents. For developers: how to integrate into pipelines. Read details.

Codex in Claude Code: OpenAI plugin for integration

Open-source OpenAI plugin integrates Codex into Claude Code. Slash commands for code review, stress tests, and background tasks. Install and test for hybrid AI workflows. Suitable for middle/senior developers.

A-Evolve: evolution of AI agents in 3 lines of code

A-Evolve Framework from Amazon automates optimization of AI agents: prompts, skills, memory. +15 p.p. on benchmarks. Integrate into projects — launch without manual setup. For LLM developers.

ToolBench: quality crisis of MCP servers

ToolBench analyzed 41K MCP servers: 76% F rating. Evaluation criteria, API barriers. Study the benchmark for AI agents and choose reliable tools.

Mozilla Mila Partnership: Open AI with Privacy

Mozilla and Mila are developing open AI with private memory architectures for agents. Learn about the cq platform and filling gaps in the stack. For IT specialists: details and prospects.

cq Mozilla: Stack Overflow for AI agents

Learn how the cq platform allows AI agents to share knowledge on API, CI/CD, and frameworks. Save tokens and resources in development. Detailed overview of the prototype for middle/senior dev.

ARC-AGI-3 benchmark: AI at 0.26% vs 100% humans

ARC-AGI-3 Breakdown: interactive tests for agent intelligence. GPT-5.4, Gemini results, comparison with humans. Study the RHAE metric and reasons for LLM failure. For AI developers.

AI agents edit Figma: MCP server in beta

AI agents now create designs in Figma using components and skills. Self-correction, integrations with IDE. Test the beta MCP server for layout automation.

Tokenmaxxing: $150K on Claude Code per month

Learn how developers spend billions of tokens on agentic AI. OpenAI leaderboards, quality risks, and Anthropic revenue growth. For middle/senior dev.

AI Agents WordPress.com: Post Creation and SEO

Learn how AI agents on WordPress.com automate content, comments, and SEO via MCP. Setup instructions for developers. Start using it right now.

Invariant Design of AI Agents for Stability

Learn how to balance flexibility and control in AI agents with invariants. Examples from Claude Code, finite automata, tools. Create reliable agents without errors. Practical tips for developers.

AI agents in commerce: automation practice

Learn how AI agents replace analysts and task trackers in the commercial department. Cases with Yandex.Metrica, meeting transcription. Save time — implement now.

Agent loop: why AI agents break in prod

Breakdown of 10 000 agent loop steps: 37% errors, 65% overhead, stack changes. Structured architecture reduces failures 17x. For middle/senior dev.

Skate: AI agent manages kanban board

Integration of AI agents with Mattermost Boards via CLI Skate. Automation of statuses, timers, task translations. Learn workflow for middle/senior dev.

React Test Automation: 7 AI Agents

Learn how a multi-agent pipeline on Claude Code migrates tests from Enzyme to RTL. Mutations, review, integration into Git. For middle/senior dev.

MCP and AI agent overload: 3 patterns

Breakdown of context crisis in MCP: why agents dumb down from extra tools. Dynamic search, Agent Skills, CLI wrappers. Reducing load by up to 90% for middle/senior dev.

BPMN environment for AI agents: 6 strategies

Learn how to adapt BPMN processes for AI agents: manifests, semantics, tests, CI/CD. Avoid semantic errors in diagrams. Prepare the environment for middle/senior developers.

NanoClaw and NanoBot: failures during testing

Analysis of deployment issues with NanoClaw and NanoBot: integrations, hallucinations, dependencies. Experience for AI agent developers. Test it yourself on VPS.

Managing AI Systems in LLM Production

Learn how to deploy LLM in production: from process analysis to AIOps and evals. Practical steps for middle/senior developers. Build reliable systems.

Renga API and AI Agents for BIM

Integrate AI with Renga API via COM and Python. Automate plugins, STDL families. Comparison with Revit, code examples. Start development without compilation.

AI Agent Failures in Production: RAG Traps

Breaking down why AI agents break in real systems: RAG, connectors, polling. Architectural fixes, tests, and cost control. For middle/senior dev.

AI agents for financial reporting without code

Learn how a chain of 17 AI agents automated B2B platform reporting. SQL, Sheets API, verification. Savings 60 h/month. Instructions for middle/senior.

Ralph loop patterns: verifier and evolution

Breakdown of 5 Ralph loop patterns for AI agents: from same-prompt to self-evolving. Verifiers, oracles, artifacts. Checklist for developers. Dive into the architecture.

Agent-Bridge for IT: canvas of AI agents sessions

Manage terminal sessions and AI agents on multiple nodes via the Agent-Bridge browser canvas. Incus isolation, real-time statuses. Deploy a self-hosted solution for middle/senior dev.

project-graph-mcp: graph for AI navigation

project-graph-mcp server builds a compressed code dependency graph for JS/TS/Python/Go. Metrics, Health Score, test checklists. Integration with agent pools. Learn how to set up.

AGI CPU from Arm: server chip for AI agents

Arm introduces AGI CPU with 136 Neoverse V3 cores for AI agents. 2× performance per rack, TDP 300 W, customers Meta and OpenAI. Details of specs and market — read the analysis.

Multi-agent system on Claude Code: deployment

How to deploy autonomous AI agents with Redis Streams bus and delegation. Practice from a single agent to a village of 12 roles. Integrations and management lessons for developers.

AI agents in projects: tests instead of review

Learn how to manage multiple projects with Claude AI through tests and dry run. Examples for EVM bots on Rust, RPC optimizations. Practice for senior developers.