Local AI pipelines
Ollama + Python pipelines for text extraction, summarization, and content processing. Nothing leaves your machine.
Custom AI workflows
Custom AI workflows, built on your machine
Local Ollama pipelines, Codex automation, and small paid apps. No cloud lock-in. No subscription fees. You own the code.
01 / Services
What I build
Codex-driven automation
Multi-step agent workflows that read, write, and execute against your files and APIs. Documented, versioned, recoverable.
Small paid web apps
One-page tools with a real use case — invoice generators, dashboards, internal admin panels. Shipped in 2-3 weeks.
02 / Process
How it works
-
Discovery call (30 min)
We identify one repetitive task worth automating.
-
Scoped proposal
Fixed price, fixed timeline, clear deliverable. No hourly.
-
Build (1-3 weeks)
Weekly demos. You see progress, not promises.
-
Handoff
Full source, README, deploy instructions. You own it.
03 / Tools
The stack
| LLM runtime | Ollama (local) or cloud APIs on request |
|---|---|
| Orchestration | Python + LangChain / n8n / Codex CLI |
| Deployment | Docker on your hardware, GitHub Pages, or your VPS |
| Data | SQLite or Postgres, your choice |
| Hosting | 100% on your infrastructure or mine, no vendor lock-in |
04 / Pricing
Pricing
Audit
$150
2-hour audit + written report + 3 prioritized automation opportunities
Build
$600+
One workflow built end-to-end. Scoped per project.
Retainer
$400/mo
Ongoing maintenance, monitoring, and iterations on one existing workflow
05 / Questions
FAQ
Do I need to know how to code?
No. I hand off working systems with documentation.
Where does the data live?
On your machine, unless you request a hosted setup.
What if I already have n8n set up?
I work with your existing stack.
What's a realistic first project?
Smallest possible scope: one workflow, one clear outcome, 1-2 weeks.