Labs
Building a data flywheel for our agent harness
An observability pipeline that reads production traces at corpus scale, names recurring failure modes, and proves the fix held.
Aug 22, 2026 · 7 min read
September 10, 2026 7 articles
Engineering
Deep dives into the infrastructure, architecture, and technical decisions behind Conversion.
Labs
Using semantic benchmarks to build a self-improving text-to-query agent
Our text-to-query agent went from 45 seconds on frontier models to 2 seconds on GLM 5.3 Flash, with the same accuracy at 1/20 the cost. This post describes the architecture, evaluation loop, and lessons learned.
Levi Kline September 2, 2026 10 min read
Engineering (6)
Labs
What we took out of the context window
A median request carried 150,000 tokens. Most of our context work since comes down to one move, applied repeatedly.
Aug 14, 2026 · 7 min read
Labs
From an agent harness to a research platform
How and why we made our harness model-agnostic, and what per-turn portability actually cost us.
Aug 6, 2026 · 7 min read
Engineering
Workflows at scale: how Conversion powers marketing automation
We’re sharing how we power marketing automation at scale using Temporal. By rethinking how workflows run, update, and scale, we’ve built a system that reliably handles millions of contacts and long-lived automations without breaking a sweat.
Nov 7, 2025 · 5 min read
Engineering
Behind the curtain: how Conversion syncs millions of customer records a day
This article dives into how we designed and built the data connector and data synchronization platform — the component responsible for syncing millions of our customers’ most important business records in from and out to their CRM’s every day.
Oct 22, 2025 · 4 min read
Engineering
Service framework: the journey behind service-to-service communication at Conversion
How we built the service-to-service communication layer behind Conversion, handling over 30 million requests a day across 350+ endpoints.
Oct 22, 2025 · 4 min read