Egor Burlakov
••7 min readA Few Things I've Learned About AI-Assisted Development in Mid-2026
Practical lessons from using coding agents on a growing codebase: repository rules, changelogs, scripts, testing, and context.
Read article
Practical lessons from using coding agents on a growing codebase: repository rules, changelogs, scripts, testing, and context.
Modern DataTools turns continuously collected data about 300 data and AI tools into comparisons, market intelligence, and evidence-backed stack recommendations.
The vector database gold rush is in full swing. But for most teams, pgvector is all you need. Here's how to decide.
Monthly analysis of the data tools gaining adoption, community interest, and reader momentum.
Every vendor now has an 'AI agent.' After evaluating dozens of them, here's what actually works and what's still a demo-only fantasy.
Three AI agents worth building on top of a modern data stack: analytics, incident triage, and internal knowledge assistants.
How to add AI on top of a modern data stack: the practical layers, tools, and guardrails for building useful agents, retrieval, and internal copilots.
Everyone's obsessed with AI models. After building a 500-tool data directory, we learned the model is 10% of the work. The other 90% is data collection, validation, and quality — and that's the real moat.
Local LLMs are practical for content generation, legal document processing, and internal knowledge bases. I benchmarked five Qwen models on my MacBook Pro. Qwen 3 14B scored 91/100 avg vs 62 for Qwen 2.5 14B -- same size, dramatically better. Newer models performed worse.
A practical 2026 guide to starting an AI-assisted software project — tools, agent orchestration, Git rules, baselining, documentation, and lessons learned.
How specs make AI coding reliable—and redefine the manager's role
Tech front lines to AI era: real stories on leading teams, testing tools, and data engineering wins—short, human-crafted lessons for your daily grind.