JetBrains reports revenue growth, net financial loss for 2025

Original Article ↗ Hacker News Discussion ↗

JetBrains’ latest financials show rising revenue but a net loss, prompting speculation that heavy spending on AI tooling and other investments is eroding profitability even as the core IDE business remains strong. Commenters are sharply divided on the company’s strategic response to AI: some say bloated products, confusing AI offerings (like Junie and Air), and declining quality are driving developers toward VS Code, terminal-based workflows, and standalone agents, while others argue JetBrains still delivers unmatched language intelligence and tooling in many ecosystems. Underneath is a broader anxiety that traditional, full-featured IDEs may be losing relevance as AI-assisted and agentic development reshape how software is written and reviewed.

Financial results and causes

  • Revenue is still growing (single-digit at the Czech entity, higher at group level), so commenters reject a simple “collapse in demand” narrative.
  • Net loss is widely attributed to:
    • Large negative “cash flow from investing” and higher staff costs.
    • A big FX hit (Czech koruna vs foreign currencies) visible in statutory filings, plus a weaker but still positive operating profit.
  • Some argue the numbers mainly show heavy reinvestment and accounting effects, not existential distress; others see it as early trouble if margins keep shrinking.

AI investment and strategy

  • Many assume the big spend is on AI: hosting models, tokens for users, and building products like Junie and Air.
  • A recurring criticism: JetBrains tried to own the whole AI stack (their own agents, curated models, extra subscriptions) instead of being a thin, flexible harness over popular models and APIs.
  • Several users say Junie / JetBrains AI felt slow, confusing, or underpowered vs Claude Code, Cursor, Codex, etc.

Product quality, performance, and UX

  • Long‑time customers report declining quality: slower IDEs, regressions (especially TypeScript, Python, remote dev, HiDPI, SSH/containers), flaky indexing/caches, and intrusive AI UI.
  • Others counter that for many languages (Java, Kotlin, .NET, databases) JetBrains IDEs remain best-in-class, with strong refactoring, navigation, and Git tooling.
  • Discontent with recent UI redesigns and “bloat” is common; some rely on legacy UI plugins.

Competition: VS Code, other IDEs, and AI tools

  • Strong split:
    • Pro‑JetBrains users see VS Code as a “fancy editor” needing many fragile extensions; JetBrains as integrated, industrial‑grade IDEs.
    • Others say VS Code (or VSCodium) is fast, good enough with LSPs and AI, and free—especially compelling for TypeScript and polyglot work.
  • Other tools mentioned: Visual Studio, Xcode, Eclipse/NetBeans, Zed, Sublime, Neovim, Emacs, DataGrip alternatives, Buildkite vs TeamCity.

Shifting workflows and “agentic” development

  • Many commenters now:
    • Use Claude/Codex/Cursor/Herdr/Pi/etc. as primary “coders”.
    • Review only diffs in GitHub/GitLab or simple editors, often skipping traditional IDEs for weeks.
    • Build custom AI orchestrators, gateways, or ADEs; some run everything from mobile or terminals.
  • Others still live in JetBrains or Visual Studio but integrate AI via terminals or plugins.

Debate over quality and safety

  • One camp claims LLM‑generated code, even with mediocre review, often beats typical human output, especially for “script‑kiddie‑level” or boilerplate tasks.
  • Another camp is alarmed: reports of huge, unreadable diffs, architecture erosion, verbose “slop”, hidden bugs, compliance/security concerns, and atrophying skills.
  • Strong disagreement over how much code must still be read and understood by humans, especially in complex or regulated systems.

Future of IDEs and JetBrains’ options

  • Some predict heavy IDEs and manual coding are on a steep decline; IDEs must evolve into AI/agent control planes (diff review, worktrees, multi‑agent orchestration, MCP/LSP bridges).
  • Others argue there will remain a sizable niche that values deterministic, offline, privacy‑respecting, non‑LLM‑centric tooling.
  • Suggested paths for JetBrains:
    • Refocus on core strengths: performance, stability, language intelligence, remote dev.
    • Offer first‑class “bring your own model/harness” integration instead of pushing proprietary AI.
    • Possibly specialize in high‑end, human‑in‑the‑loop workflows (code review, debugging, refactoring) rather than compete head‑on with AI‑native IDE startups.