Blog

Insights, tutorials, and updates from the Moss team.

What Happens When You Remove the Network Hop from RAG

March 17, 2026·Sri Raghu Malireddi & Harsha Nalluru·11 min read

What Happens When You Remove the Network Hop from RAG

We profiled a production RAG pipeline before and after eliminating the retrieval network round-trip. The results explain why your AI agent feels slower than it should.

The Retrieval Latency Tax: Why Your AI Agent Feels Slow (And It's Not the LLM)

March 17, 2026·Sri Raghu Malireddi & Harsha Nalluru·11 min read

The Retrieval Latency Tax: Why Your AI Agent Feels Slow (And It's Not the LLM)

Everyone blames the model. But in real-time AI, the real bottleneck is retrieval. Here's the data that proves it, and what it means for the next generation of AI applications.

March 10, 2026·Sri Raghu Malireddi & Harsha Nalluru·7 min read

We Spent a Decade Making AI Feel Instant. Here's What We Learned.

The founders of Moss share why they left Microsoft and Grammarly to build the real-time retrieval runtime, and why the future of conversational AI depends on killing latency at the source.

Loading

Preparing your content