---
title: "Three Years of Building Agents in Production (Part 2)"
url: https://daily.dev/posts/three-years-of-building-agents-in-production-part-2--9uvhtnda1
source_url: https://www.mabl.com/blog/three-years-of-building-agents-in-production-part-2
type: article
source: "Mabl Engineering Blog"
published: 2026-07-08T15:00:46.474Z
updated: 2026-07-08T16:21:44.455Z
tags: ["testing", "llm", "ai-agents", "rag"]
reading_time: 11
upvotes: 2
comments: 1
language: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Three Years of Building Agents in Production (Part 2)

**[Mabl Engineering Blog](https://daily.dev/sources/mabl)** · 11 min read · 2 upvotes · 1 comments

## Summary

A mabl engineer shares lessons from three years building AI agents in production for software testing. After a successful demo exposed infrastructure fragility, the team paused feature development for a winter sprint to build a unified Agent Framework. The framework rests on three pillars: shared context compaction strategies to manage LLM token limits, composable shared tool definitions with a message handling layer that catches hallucinations and enforces self-correction loops, and evals-as-code with LLM-as-a-judge suites to prevent regressions. The result is an ecosystem of six agents (authoring, recovery, results analysis, failure diagnosis, application modeling, session review) that can be assembled rather than hand-crafted. Key insight: scaling AI is not about better prompts but about returning to foundational software engineering practices — reusability, error handling, and standardized testing become more critical, not less, when non-deterministic tools are involved.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.mabl.com/blog/three-years-of-building-agents-in-production-part-2>

## Community discussion

Top comments from developers on daily.dev.

**@kartiknvj** · 1 upvotes

> The move to evals-as-code is the part I would underline for anyone starting this, because it turns agent quality into something you can diff and review in a PR instead of a vibe check. Pairing that with LLM-as-judge suites only works once you have pinned the judge against human labels, otherwise you are trusting one model's opinion of another. With six specialized agents, are you running the judge suites per-agent, or against the end-to-end trajectory where the handoffs tend to break?

## Similar posts on daily.dev

- [Three Years of Building Agents in Production \(Part 1\)](https://daily.dev/posts/three-years-of-building-agents-in-production-part-1--d2wjasjzx) · Mabl Engineering Blog · 3 upvotes · 1 comments
- [From prototype to production: lessons learned taking AI-built software to enterprise scale](https://daily.dev/posts/from-prototype-to-production-lessons-learned-taking-ai-built-software-to-enterprise-scale-kbmv5vowg) · Atlassian · 0 upvotes · 0 comments

---

Tags: [#testing](https://daily.dev/tags/testing), [#llm](https://daily.dev/tags/llm), [#ai-agents](https://daily.dev/tags/ai-agents), [#rag](https://daily.dev/tags/rag)

[View this post on daily.dev](https://daily.dev/posts/three-years-of-building-agents-in-production-part-2--9uvhtnda1)