Best tech stacks for agentic coding: training data volume decides everything
aiblastoff · reddit · 2026-10-03
A Reddit deep-dive on choosing tech stacks in the agentic coding era. The author's core thesis: the language matters less than training data volume — stacks where the model has seen every mistake and fix (NextJS, Python) work best, while rarer languages invite hallucinated APIs or outdated code.
Current recommendations:
- Frontend: plain HTML/CSS/JS or NextJS + TypeScript (most training data)
- Backend: Python, Node, or Go; Python for AI/data-heavy backends
- Small fast services/APIs: Go
- Performance-critical or games: Rust (agent wrestles the compiler)
- Mobile: React Native (shares ecosystem with web)
- Enterprise: stick with Java/C/C++ (huge legacy codebases)
- WebAssembly: hot paths only, not whole apps
The author personally uses plain HTML/CSS/JS on cheap shared hosting, PHP when a backend is needed, and NextJS + Python/Node on a VPS — noting NextJS's built-in SEO features as a plus.
More from coding & agent
- Dev turns a Pixel 10 Pro XL into an offline OpenAI-compatible server running Gemma 4 E4B at ~11 tok/s — VerityAISolutions · 2026-10-03
- NVIDIA hands over first Vera CPU to test AI agent code environments — denisyarats · 2026-10-03
- TL: a token-efficient language that cuts LLM code tokens by 30-60% — Ok-Condition7148 · 2026-10-03
- Context compaction backfires: Billion Context plugin sends coding agent into a loop — Iory1998 · 2026-10-03
- A weekend, a few hundred lines: building a personal AI voice agent with Telnyx, gpt-live and Cloudflare — itsOmSarraf_ · 2026-10-03
- Cloudflare integrates Pi Durable into its agents SDK, and it just works — irvinebroque · 2026-10-03