Tag
All blog posts with this tag.
AI already writes working functions, but a product is architecture, context, risk, and accountability. The boundary between code and engineering, a 15-year legacy case study, and developer skills in 2026.
Millions of lines, hundreds of tables, dozens of integrations: what AI grasps in hours, where it fails, how teams deploy RAG and repo indexing, and why expert + model beats either alone.
Habr case study: from Vanessa Automation to native TestClient in Python — where Codex stalled for a week, Fable 5 mapped the protocol in a day for qa-mcp.
Fetch JSON transcripts with timestamps via a RapidAPI endpoint — for LLM summarization, vector search, and subtitles without brittle page parsing.
Shumai offers media storage, frame-accurate review, collaboration, and an AI agent with semantic search. Deploy in minutes with Docker Compose or npm.
Flowcharts, UML, BPMN, network topologies, and CAD drawings: why PNG ≠ diagram, what OCR, vectorization, CAD, computer vision, and VLMs deliver in 2025–2026, and which pipeline actually works.
The 2025–2026 content race: why mass AI generation fails to scale linearly, how search engines judge whole sites, and what works instead of content farms.
A home chatbot experiment: fine-tune a tiny local LLM with Unsloth and two-letter category codes instead of label names for metadata-aware RAG.
Practical guide to NeuralBridge SDK: multi-provider failover, cascade self-healing, and observability for OpenAI-compatible LLM calls in Python.
80M+ classification rows across 40M public repos — README, issue, and PR language signals under CC0 for evaluating multilingual developer AI tools.
A senior dev's honest take: Agent = Model + Scaffold, context engineering, the factory model, and why vibe coding does not scale past prototypes.
Google Apps Script, hidden Drive OCR, Gemini key rotation, and LockService — a Habr pipeline for cataloging huge archives.
Platform9, Monday.com, and PlayStation on restructuring the SDLC around AI, orchestration engineers, and review bottlenecks.
Head-to-head on schema and instruction tasks: DeepSeek 38–33, one regex vs split patterns, fewer unsolicited extras.
One input, weight and bias, training epochs, and a linear decision boundary — neural network fundamentals without frameworks.
Google Genkit for Go—structured outputs, multi-step flows, tools, observability, and model swaps without rewriting your logic.
INSAIT released Gemma 3–based 12B and 27B models with Ukrainian support, multimodal features, and infrastructure hosted in Ukraine.
Dev.to benchmark: ~0.9s vs ~3.2s p99, order-of-magnitude API economics, and routing most traffic to a cheap model with premium fallback.
40% of teams cite security as the top barrier to scaling agentic AI. Docker Sandboxes, Engine hardening, and governance for autonomous code.
Million-token context windows still move JSON between planner, tools, and memory. ULMEN LLM claims 44% fewer tokens in the author's tests.
Natural language to SQL via LLM, in-process DuckDB, a three-stage pipeline, and read-only guardrails for safe analytics.
arXiv preprint: a KL-divergence metric between prior and posterior finds OOD patches in inverse problems without calibration data on shifted domains.
Metrics Layer / Headless BI as an executable contract for metrics—not a data catalog—and why LLMs should not invent SQL on their own.
Same question, different chart data—surfacing when models answered from priors instead of visual reasoning.
Prompt pools, bulk runs, SoV, Sentiment, and Accuracy for Generative Engine Optimization. Q2 2026.
A ReAct agent with tools beats general chatbots on merger/case tasks—RAG, web fallback, and clarifying questions.
ROUGE and BERTScore can score opposite texts alike. MATCHA adds counterfactual contradiction penalties—up to ~20% better human agreement.
Your own ChatGPT-style UI for Ollama and OpenAI-compatible APIs — a step-by-step deploy with Let's Encrypt.
Gaze coherence between people in a scene is a semantic detection axis orthogonal to pixels—+3.7 pp on COCOAI Interaction in the paper.
An honest take on llms.txt in AISO: agent site map vs GEO citation myths. Q2 2026.
Practical guide to robots.txt for AI crawlers: GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot, Google vs Yandex split, Allow/Disallow, post-deploy checklist, and AISO linkage. Q2 2026.
Whisper + LLM summary breaks in real practice: healthcare needs clinical reasoning engines and EMR integration, not plain text.
Google I/O Writing Challenge: multi-agent orchestration assembled an OS core with no hand-written code — and shifts the developer role.
Chroma, Pinecone, Qdrant, Weaviate—not QPS charts, but packaging, API stability, and the path from dev to production.
Copilot/Cursor code compiles but breaks in the browser — why to run the app and Playwright/Cypress in one compose stack.
LanguageModel.create() with no server or API key—limits, ~4K context, and why a hosted LLM fallback is mandatory.
RankCaster AI built an autonomous tester with Claude Code, agent-browser, and read-only psql — regression dropped from 48 hours to 40 minutes.