Skip to content

Alexander Bobkov

Frontend Tech Lead · Symfio

Recent articles

  • March 23, 2026

    Agentic UI: Building Frontend Interfaces for Multi-Step AI Agent Workflows

    Single-turn LLM calls are a solved UI problem. Agent workflows are not. Tool call progress streaming, human-in-the-loop confirmations, mid-agent cancellation, and how not to mislead users about what the AI is doing.

  • March 11, 2026

    Building Maintainable Design Systems in 2026: From Tokens to Semantic Layers and AI-Assisted Theming

    Why Figma → tokens → CSS vars is no longer enough. Introducing semantic tokens, code-level variants, and LLM-driven generative theming into a real design system.

  • February 24, 2026

    Web Vitals in 2026: Keeping LCP Under 1.5s With 500k+ Users and Heavy SSR

    Real-world cases: font loading, hero images, third-party scripts killing INP, partial prerendering, and edge caching tricks to keep LCP under 1.5s at scale.

  • February 6, 2026

    Micro-Frontends in 2026: When Module Federation Still Works — and When to Add Web Components

    Comparing Module Federation, single-spa, MF v2 — when to add Web Components, and how to migrate a legacy frontend while shipping new features in parallel.

  • January 11, 2026

    GDPR + AI in Frontend: Safe LLM Integration Without PII Leaks in 2026

    In 2026, adding smart UI features demands careful PII handling. How to safely integrate LLMs into web apps without risking user privacy or GDPR compliance violations.

  • December 9, 2025

    Frontend Testing Strategy at Scale: What to Test, What to Skip, and Why

    The testing pyramid applied to React and RSC at scale: what belongs in unit vs integration vs E2E, visual regression testing, testing AI streaming features, and getting large teams to actually write tests.

  • November 2, 2025

    Systems Thinking in Website Builder and Multi-Site Architecture

    The nuances of multi-tenant system design that rarely get discussed openly — yet without them, Website Builder and multisite architectures inevitably fall apart.

  • October 14, 2025

    How I Run Frontend Architecture Decisions: RFCs, ADRs, and Getting Team Buy-In

    A lightweight RFC and ADR process that frontend teams actually use: how to write proposals engineers read, run async reviews that do not die in Slack, and handle disagreement without a culture war.

  • September 13, 2025

    React Server Components + Streaming in Production: A Deep Dive (App Router 2025–2026)

    How we migrated 400+ pages to RSC, the trade-offs that surfaced, how to measure and fix waterfalls, and why Suspense boundaries are an engineering choice, not magic.

  • July 29, 2025

    Developer Experience at Scale: Monorepo Tooling, Build Speed, and Shared Configs

    How to set up a frontend monorepo that engineers enjoy working in: Turborepo vs Nx trade-offs, remote caching, enforcing module boundaries, and shared ESLint and TypeScript configs across workspaces.

  • June 24, 2025

    Frontend Security Beyond GDPR: CSP, XSS, Dependency Auditing, and Supply Chain Risk

    Practical frontend security defaults every tech lead should configure: Content Security Policy, XSS prevention in React, npm supply chain risk, and dependency auditing in CI pipelines.

  • May 18, 2025

    Accessibility as an Engineering Discipline: From ARIA Live Regions to axe-core in CI

    How we raised an a11y score from 60% to 98% in a 5M MAU product — without team friction. Custom React hooks, ARIA live regions, semantic HTML, and axe-core in CI.

  • April 14, 2025

    TypeScript in Large Codebases: Branded Types, Infer, Module Augmentation, and tRPC-Style Safety

    Escaping "any hell" in large codebases: branded types, const assertions, module augmentation, and tRPC-style end-to-end type safety. Why zod + ts-pattern is the real solution.

  • March 8, 2025

    Integrating LLMs Into Frontend: Streaming Responses, Optimistic UI, and Error Handling in React

    Embedding AI chat and content generation in React: streaming responses, optimistic UI, abort controllers, rate limiting, and keeping Core Web Vitals intact under LLM load.

© 2026 Alexander Bobkov GitHub RSS