The Most Stable Architecture for Enterprise Agents
amitavital · reddit · 2026-07-15
The author reviews their process of building agentic systems within a company, concluding that a single, all-encompassing agent typically fails.
They encountered several pitfalls along the way:
- Stuffing all company knowledge into a giant system prompt works temporarily but quickly spirals out of control and becomes hard to debug.
- Giving one agent 400 tools makes it harder for the model to choose the right tool, leading to chaotic combinations.
- Having the main agent dynamically spawn numerous sub-agents ultimately results in unclear responsibilities, context leakage, and untraceable answers.
The effective structure they found is:
- A top-level orchestrator dedicated solely to routing, determining which business domain a request belongs to.
- Each sub-agent covers only one domain and is equipped with a small, dedicated set of tools.
- For example, a billing agent only handles invoices, and an orders agent only touches orders, preventing cross-contamination.
The author believes this architecture leads to more accurate routing, smaller context windows, and easier debugging. Adding new capabilities simply means adding a new sub-agent, without rewriting the entire system.
More from coding & agent
- Soft Clamp cuts tool-call overuse in multi-teacher distillation, from 13.7% to 9.0% — antgroup · 2026-07-21
- Agent harness memory loss and compaction are still a major usability problem — adityaag · 2026-07-21
- SpecJudge runs locally on Ollama to pick the right-sized AI model for your project — jokiruiz · 2026-07-21
- A developer maps out six design rules for CLIs that humans and AI agents can both use — yujiezha · 2026-07-21
- A coding-agent skill that forces ADHD-friendly, answer-first output — ayghri · 2026-07-21
- A set of agent skills for CAD, robotics, and hardware design — earthtojake · 2026-07-21