<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Neul Labs Blog</title><description>High-performance, Rust-accelerated infrastructure for AI agents. Deep dives, benchmarks, and engineering notes from the neul-labs org.</description><link>https://www.neullabs.com/</link><language>en-gb</language><item><title>Notes from Shipping 28 Open-Source Projects in One Year</title><link>https://www.neullabs.com/blog/notes-from-shipping-28-projects/</link><guid isPermaLink="true">https://www.neullabs.com/blog/notes-from-shipping-28-projects/</guid><description>What we learned shipping 28 MIT-licensed projects under the neul-labs org — across agent runtimes, Rust accelerators, dev tools and governance. The themes that actually emerged, and the ones we expected but didn&apos;t.</description><pubDate>Fri, 05 Jun 2026 00:00:00 GMT</pubDate><category>meta</category><category>open-source</category><category>neul-labs</category><category>agentic-infrastructure</category><author>Dipankar Sarkar</author></item><item><title>Drop-in Compatibility as a Design Principle</title><link>https://www.neullabs.com/blog/drop-in-compatibility-as-a-design-principle/</link><guid isPermaLink="true">https://www.neullabs.com/blog/drop-in-compatibility-as-a-design-principle/</guid><description>Why most of the Neul Labs portfolio — rjest, rpytest, rninja, recurl, rewget, stout, fast-litellm, fast-langgraph, fast-crewai, fast-axolotl — refuses to change your config, your CLI, or your imports. The rule, the trade-offs, and the cases where we break it.</description><pubDate>Fri, 05 Jun 2026 00:00:00 GMT</pubDate><category>design</category><category>drop-in</category><category>rjest</category><category>rpytest</category><category>rninja</category><category>fast-litellm</category><category>fast-langgraph</category><category>open-source</category><author>Dipankar Sarkar</author></item><item><title>The Agentic Infrastructure Stack: What We&apos;re Building and Why</title><link>https://www.neullabs.com/blog/agentic-infrastructure-stack/</link><guid isPermaLink="true">https://www.neullabs.com/blog/agentic-infrastructure-stack/</guid><description>A reference architecture for an agentic stack — workspace runtime, memory, ORM-level policy, orchestration, payments — built from the Neul Labs primitives: agentvfs, brat, fastagentic, ormai, mcp-pay and memorg. With pointers to each repo and the open problems we still see.</description><pubDate>Fri, 05 Jun 2026 00:00:00 GMT</pubDate><category>agents</category><category>architecture</category><category>agentvfs</category><category>brat</category><category>ormai</category><category>mcp-pay</category><category>memorg</category><category>fastagentic</category><category>agentic-infrastructure</category><author>Dipankar Sarkar</author></item><item><title>Why We Chose Rust to Accelerate Python AI Infrastructure</title><link>https://www.neullabs.com/blog/why-rust-for-ai-infrastructure/</link><guid isPermaLink="true">https://www.neullabs.com/blog/why-rust-for-ai-infrastructure/</guid><description>Python&apos;s GIL and single-threaded ORMs create bottlenecks in production AI agents. Rust + PyO3 delivers 3x-700x faster drop-in replacements for LiteLLM, LangGraph, and CrewAI with zero code changes.</description><pubDate>Sat, 04 Apr 2026 00:00:00 GMT</pubDate><category>rust</category><category>python</category><category>performance</category><category>pyo3</category><category>ai-agents</category><author>Dipankar Sarkar</author></item><item><title>How to Make LangGraph Checkpointing 700x Faster with Rust</title><link>https://www.neullabs.com/blog/langgraph-checkpoint-performance/</link><guid isPermaLink="true">https://www.neullabs.com/blog/langgraph-checkpoint-performance/</guid><description>LangGraph checkpoint serialization is the #1 performance bottleneck in production agent deployments. fast-langgraph uses Rust to achieve 737x faster serialization, 151x faster deserialization, and 2.8x faster end-to-end execution.</description><pubDate>Sat, 04 Apr 2026 00:00:00 GMT</pubDate><category>langgraph</category><category>benchmarks</category><category>rust</category><category>performance</category><category>checkpointing</category><author>Dipankar Sarkar</author></item><item><title>How to Run Multiple AI Coding Agents in Parallel with brat</title><link>https://www.neullabs.com/blog/multi-agent-orchestration-with-brat/</link><guid isPermaLink="true">https://www.neullabs.com/blog/multi-agent-orchestration-with-brat/</guid><description>brat is an open-source multi-agent orchestration harness that coordinates Claude Code, Aider, Codex, and other AI coding tools on shared codebases with crash-safe state, merge queues, and a real-time dashboard.</description><pubDate>Sat, 04 Apr 2026 00:00:00 GMT</pubDate><category>agents</category><category>orchestration</category><category>brat</category><category>claude-code</category><category>aider</category><category>codex</category><author>Dipankar Sarkar</author></item><item><title>Python Task Queue Without Redis: How FastWorker Replaces Celery + Redis</title><link>https://www.neullabs.com/blog/brokerless-task-queues-python/</link><guid isPermaLink="true">https://www.neullabs.com/blog/brokerless-task-queues-python/</guid><description>FastWorker is a brokerless Python task queue that eliminates Redis and RabbitMQ. It supports priority queues, auto-discovery, distributed workers, and a built-in monitoring dashboard — all without an external message broker.</description><pubDate>Sat, 04 Apr 2026 00:00:00 GMT</pubDate><category>python</category><category>fastworker</category><category>celery-alternative</category><category>task-queue</category><category>redis</category><author>Dipankar Sarkar</author></item></channel></rss>