Venture Crane

infrastructure

Articles

Overlay Drift: The Fleet Observability Gap That Looks Like a Deployment

Jun 17, 2026 · 5 min read

Deploying a new overlay version does not mean all instances have it - persistent processes require active drift detection, not deployment confirmation.

Live Reconfiguration for Persistent AI Agent Processes

Jun 16, 2026 · 5 min read

How to apply configuration changes to a running AI agent without downtime, and why "durable" apply semantics matter more than the detection mechanism.

Read Your Own Codebase First

Jun 15, 2026 · 5 min read

Prior implementations encode constraints vendor docs omit - check your codebase before any setup guide for a library you have integrated before.

You Cannot Instruct an Agent Not to Leak

May 27, 2026 · 8 min read

Memory notes told the agent not to echo secret values. It did anyway, twice. The fix was structural enforcement at the tool boundary, not better instructions.

Observability That Is Born and Dies With the Tenant

May 25, 2026 · 8 min read

When every customer runs an isolated agent deployment, monitoring becomes per-tenant and lifecycle-aware. The heartbeat, webhook, and fleet stack we wired.

Retiring a Fork That Stopped Earning Its Keep

May 23, 2026 · 8 min read

A dependency you control "for safety" can quietly become pure cost. We audited a fork from first principles and found the safe move was to delete it.

Migrating an Agent Fleet from Tarball Installs to a Private npm Registry

Apr 22, 2026 · 6 min read

12 critical Dependabot alerts, one incomplete migration, and four PRs to finish the job. What the move from tarball URLs to GitHub Packages actually looks like.

Tool Registration Is Not Tool Integration

Mar 28, 2026 · 8 min read

Registering an MCP server in a CLI config means the CLI can discover the tools. It does not mean the tools can access the credentials they need to function.

Taking Product Development Offline with Local LLMs

Mar 2, 2026 · 9 min read

We set up four specialized local models on a fanless laptop for offline product work. Here is what we built, what works, and what we are measuring.

Where We Stand: AI Agent Operations in February 2026

Feb 21, 2026 · 15 min read

An honest field report on running AI agent teams in production. What the stack looks like, what works, what breaks, and what the data says versus the marketing.

What Running Multiple Ventures with AI Agents Actually Costs

Feb 19, 2026 · 13 min read

Every line item for running an AI-native dev lab across multiple projects. Total: about $490 a month.

Secrets Management for AI Agent Teams

Feb 16, 2026 · 10 min read

How to manage secrets across projects, machines, and autonomous agents without .env files, hardcoded credentials, or accidental exposure.

How We Built an Agent Context Management System

Feb 14, 2026 · 28 min read

Building centralized context management so AI agents start every session with the right knowledge.

Documentation as Operational Infrastructure

Feb 10, 2026 · 13 min read

Why we treat runbooks, ADRs, and handoff records as infrastructure that self-heals, version-tracks, and delivers itself to agents automatically.

From Monolith to Microworker - Decommissioning the Relay

Feb 5, 2026 · 9 min read

We deleted a 3,234-line Cloudflare Worker, its database, and its storage bucket. Here is what we learned about scope creep in serverless architectures.

Staging Environments for AI Agents

Feb 3, 2026 · 10 min read

A 4-phase environment strategy for AI agent infrastructure - Cloudflare splits, automated CI/CD, scoped secrets, and agent-aware routing.

Sessions as First-Class Citizens - Heartbeats, Handoffs, and Abandoned Work

Jan 31, 2026 · 14 min read

Why we gave AI agent sessions heartbeats, idempotent handoff storage, and a full lifecycle - the same reliability patterns used in distributed systems.

Building an MCP Server for Workflow Orchestration

Jan 29, 2026 · 14 min read

A practical walkthrough of building an MCP server that bridges an AI coding CLI to a custom backend API, with lessons on tool design and fleet deployment.

Secrets Injection at Agent Launch Time

Jan 27, 2026 · 12 min read

How a CLI launcher scans repos, matches them to projects, and injects the right secrets without .env files or hardcoded credentials.

Fleet Management for One Person

Jan 24, 2026 · 9 min read

How a solo founder manages a distributed dev fleet with Tailscale, idempotent bootstrap scripts, SSH mesh networking, and macOS hardening.

One Monorepo, Multiple Ventures - Registry-Driven Multi-Tenant Infrastructure

Jan 22, 2026 · 10 min read

How a JSON venture registry with capability flags lets a single monorepo serve multiple products without infrastructure sprawl or cross-contamination.

Why We Built a Development Lab Instead of a Product

Jan 13, 2026 · 7 min read

Most founders pick one idea and go all-in. We built shared infrastructure first. Here is why.

Ship Log

Jun 23, 2026

Connector Platform, Console Correction, Durable Execution, and the Ops Handbook

Jun 21, 2026

Done Means Wired - Shipping the Verify Gate

Jun 17, 2026

Conversational Channel, Model Tiering, and Security Hardening - June Sprint

Jun 11, 2026

Giving the agent access to a managed mailbox - read and draft first

Jun 6, 2026

The launcher now self-heals a stale local tree so new skills reach agents

Jun 3, 2026

Completing the pivot to Infisical - every Bitwarden reference removed

Jun 2, 2026

Clerk SSR auth on the internal console, reusing the pattern we already proved

May 28, 2026

Moving secret-leak prevention outside the agent

May 26, 2026

Binding claude.ai projects to the right venture

May 20, 2026

Writing down every token before one of them broke

Apr 22, 2026

Semgrep as a CI gate

Apr 12, 2026

Killing the QA grading theatre

Feb 25, 2026

Fleet Sprint Reliability Improvements

Feb 22, 2026

Giving the browser advisor live context access via remote MCP

Feb 17, 2026

Three CLIs, one sprint, zero excuses left

Feb 17, 2026

Making every web app installable in one session

Feb 17, 2026

Cloudflare Makes Edge Agents Easier

Feb 16, 2026

Building a go-live checklist that tests itself

Feb 16, 2026

Vault sub-paths for secrets that should not be injected

Feb 16, 2026

Fixing RSS feed quality after an external review

Feb 16, 2026

Decomposing agent instructions into a routing table

Feb 15, 2026

Adding a newsletter with Buttondown

Feb 15, 2026

Adding a contact form without a form service

Feb 15, 2026

Staging environments for agent infrastructure

Feb 11, 2026

Decommissioning the legacy webhook worker

Feb 7, 2026

Dropping categories for tags-only knowledge taxonomy

Feb 6, 2026

Moving enterprise context from Apple Notes to D1

← All tags