<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>PlatStone</title><description>PlatStone gives development teams a private AI platform — local models and a RAG server trained on your codebase, wired into every developer’s IDE. Your code never leaves your network.</description><link>https://platstone.dev/</link><item><title>The Karpathy Method: How to Actually Build Software With LLMs</title><link>https://platstone.dev/blog/the-karpathy-method/</link><guid isPermaLink="true">https://platstone.dev/blog/the-karpathy-method/</guid><description>Andrej Karpathy popularized both &apos;vibe coding&apos; and the discipline of keeping AI on a tight leash. Here&apos;s how his generation-verification loop works in practice — and why a private, local platform is the ideal place to run it.</description><pubDate>Thu, 25 Jun 2026 00:00:00 GMT</pubDate></item><item><title>Run AI Coding Assistants Locally: Why Your Code Should Never Leave Your Network</title><link>https://platstone.dev/blog/run-ai-coding-assistants-locally/</link><guid isPermaLink="true">https://platstone.dev/blog/run-ai-coding-assistants-locally/</guid><description>Cloud AI coding tools send your source code to someone else&apos;s servers. For most serious teams, that&apos;s a risk they can&apos;t accept. Here&apos;s how local AI gives you the speed without the exposure.</description><pubDate>Mon, 22 Jun 2026 00:00:00 GMT</pubDate></item><item><title>Local RAG for Engineering Teams: Give Your AI Your Codebase, Not the Internet&apos;s</title><link>https://platstone.dev/blog/local-rag-for-engineering-teams/</link><guid isPermaLink="true">https://platstone.dev/blog/local-rag-for-engineering-teams/</guid><description>A generic model knows public code. A local RAG server knows yours. Here&apos;s how retrieval over your own repos, docs, and standards turns a good assistant into a great one — without anything leaving your network.</description><pubDate>Sat, 20 Jun 2026 00:00:00 GMT</pubDate></item><item><title>Cursor, Continue &amp; JetBrains: Connect Your Whole Team to a Private LLM</title><link>https://platstone.dev/blog/connect-your-team-to-a-private-llm/</link><guid isPermaLink="true">https://platstone.dev/blog/connect-your-team-to-a-private-llm/</guid><description>Your developers already live in their IDEs. Here&apos;s how to point Cursor, VS Code, and JetBrains at a private, self-hosted model so every engineer gets AI in their editor — without your code leaving the building.</description><pubDate>Thu, 18 Jun 2026 00:00:00 GMT</pubDate></item><item><title>Self-Hosted LLMs in 2026: Hardware, Models, and What It Actually Costs</title><link>https://platstone.dev/blog/self-hosted-llms-2026-hardware-models-cost/</link><guid isPermaLink="true">https://platstone.dev/blog/self-hosted-llms-2026-hardware-models-cost/</guid><description>A practical, no-hype guide to running your own LLMs in 2026 — what hardware you really need, which open models to consider, and how the total cost compares to cloud AI subscriptions.</description><pubDate>Sun, 14 Jun 2026 00:00:00 GMT</pubDate></item><item><title>AI Without the Compliance Nightmare: Local Models for Finance, Healthcare &amp; Legal</title><link>https://platstone.dev/blog/local-ai-for-regulated-industries/</link><guid isPermaLink="true">https://platstone.dev/blog/local-ai-for-regulated-industries/</guid><description>In regulated industries, sending source code or data to a third-party AI is a non-starter. Local, air-gapped AI lets your developers move fast while staying inside the lines. Here&apos;s how.</description><pubDate>Fri, 12 Jun 2026 00:00:00 GMT</pubDate></item><item><title>Which Open Model Should Your Dev Team Run? Llama vs. Qwen vs. DeepSeek</title><link>https://platstone.dev/blog/choosing-an-open-model-for-your-dev-team/</link><guid isPermaLink="true">https://platstone.dev/blog/choosing-an-open-model-for-your-dev-team/</guid><description>Choosing the right open model is the difference between an assistant your team loves and one they ignore. A practical framework for picking — and why the answer changes over time.</description><pubDate>Mon, 08 Jun 2026 00:00:00 GMT</pubDate></item><item><title>Cloud AI Coding Tools vs. Local: The Real 2026 Cost Comparison</title><link>https://platstone.dev/blog/cloud-ai-coding-tools-vs-local-cost/</link><guid isPermaLink="true">https://platstone.dev/blog/cloud-ai-coding-tools-vs-local-cost/</guid><description>Per-seat AI subscriptions look cheap until you do the math across a growing team. Here&apos;s an honest total-cost comparison between cloud AI coding tools and a self-hosted platform.</description><pubDate>Thu, 04 Jun 2026 00:00:00 GMT</pubDate></item><item><title>From Pilot to Every Developer: Rolling Out Private AI Across Your Org</title><link>https://platstone.dev/blog/from-pilot-to-every-developer/</link><guid isPermaLink="true">https://platstone.dev/blog/from-pilot-to-every-developer/</guid><description>A successful AI pilot with five engineers is easy. Getting all 200 to actually use it is the hard part. A practical playbook for rolling out a private AI platform org-wide.</description><pubDate>Sat, 30 May 2026 00:00:00 GMT</pubDate></item></channel></rss>