From Months to Minutes: Rebuilding Our AI Infrastructure for Scale

WRITER engineers share a deep dive into rebuilding AI infrastructure for scale. Learn how they slashed model integration time from months to minutes. The piece details how a new LLM Gateway replaces hard-coded integrations with a dynamic, self-service platform. Engineers can now add models in seconds, configure guardrails instantly, and track every request with full visibility. Learn how infrastructure was reimagined to serve thousands of agents across hundreds of companies, enabling flexible model choice, real-time supervision, and instant credential rotation—all without downtime.

Writer Team

Writer Engineering

How personalized context quietly degrades AI accuracy: a deeper look

Writer Team

Cerebro: An open source agentic system for security alert triage

Ben Popper

When too many tools become too much context

Ashley Weaver

Rethinking Security: Moving from Human Speed to Machine Speed

Avoid context rot and improve tool accuracy for AI agents using MCP

Dennis Thompson

Beyond vibe coding: prototyping enterprise agents

Sam Julien

Palmyra-mini: Small models, big throughput, powerful reasoning

Rakshith Vasudev

WRITER’s Palmyra X5 on Amazon Bedrock: Unlock long context AI for enterprise

Ashley Weaver

The living brain of the enterprise

Matan-Paul Shetrit

The orchestration graph

Matan-Paul Shetrit

Say hello to Action Agent

Waseem AlShikh

Everyone is a manager now

Matan-Paul Shetrit

Navigating the challenges of generative AI and "vendor lock-in" in enterprises

Waseem AlShikh

Introducing self-evolving models

Waseem AlShikh

Synthetic data: Busting the myths holding back enterprise AI progress

Waseem AlShikh

Short transformers: easily prune redundant LLM layers

Melisa Russak