---
title: "Create Your Own Work Stack: Deploying Self-hosted AI and Tooling in a Tech Startup"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.08116v1"
permalink: "https://feed.astralor.com/items/itm_464909b8611c45fc032e396e"
published: "2026-11-05T14:49:11.000Z"
score: 62
channel: "tech"
tags: ["AI Agent","Anthropic","推理效率","RAG","Create Your Own","Work Stack","Deploying Self-hosted","Tooling"]
---

# Create Your Own Work Stack: Deploying Self-hosted AI and Tooling in a Tech Startup

## At a glance

Self-hosting AI is often presented as a way to reduce inference costs and increase control, but deployment introduces its own costs. We report our experience deploying a hybrid AI stack at Grayhat Developers, a small so…

## Key points

- Self-hosting AI is often presented as a way to reduce inference costs and increase control, but deployment introduces its own costs.
- We report our experience deploying a hybrid AI stack at Grayhat Developers, a small software company with an AI-intensive development workflow.
- Our initial motivation was to counter the growing cost of commercial AI usage: Claude/Codex incurred approximately \$325 per month, with more than 25.93 billion recorded tokens.

## Full text

Self-hosting AI is often presented as a way to reduce inference costs and increase control, but deployment introduces its own costs. We report our experience deploying a hybrid AI stack at Grayhat Developers, a small software company with an AI-intensive development workflow. Our initial motivation was to counter the growing cost of commercial AI usage: Claude/Codex incurred approximately \$325 per month, with more than 25.93 billion recorded tokens. We experimented with local open-source models as replacements for commercial coding agents, but after learning that their performance and inference speed were insufficient, we adopted a hybrid approach in which commercial agents remained responsible for complex tasks while local models handled lower-complexity tasks, organizational information retrieval, and synchronization with self-hosted tools including Outline, Plane, and ERPNext. We estimated that approximately 30--40\% of our commercial AI workload was associated with context transfer, documentation synchronization, task management, and related operational coordination. By moving this work to local models, we estimate that approximately 90\% of this coordination workload could be handled without commercial inference, corresponding to an estimated 27--36\% reduction in overall commercial-AI workload or cost-equivalent usage. Our first internal deployment nevertheless failed to achieve adoption because it introduced UX friction. Replacing it with a centrally configured deployment substantially improved feedback and adoption. We conclude that local inference is most useful not as a wholesale replacement for commercial AI, but as part of a hybrid architecture that assigns tasks according to their capability requirements, cost, context demands, and operational sensitivity.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_464909b8611c45fc032e396e