Systems we've
built
Automation systems, internal tools, and product foundations that replace manual work and scale operations.
Every project here started with a repeated problem. Instead of patching it, we turned it into a system.
AI Content Automation System
The content system we run CodeITronics on: blog drafts, social posts for six platforms, a weekly newsletter and an n8n planner for ten formats, with a person approving everything.
Turning each idea into a blog post, six platform-specific social posts and a newsletter, every week, without the voice drifting, took more time than writing the idea down.
A website-native content system (brand voice, blog drafting, social generator, queue, newsletter) plus n8n workflows for scheduling and multi-format planning. Everything is drafted and checked automatically; nothing goes out until a person schedules it.
- 75 LinkedIn posts written and scheduled for the CodeITronics page
- A 30-part AI Systems blog series written and scheduled
- Scheduler checks for due posts 4 times a day
- Both n8n workflows open-sourced as client templates
Internal Ops Automation for Founder
Custom system to manage leads, tasks, content, and execution from a single workflow layer.
Founder was juggling multiple tools with no central system, leading to context switching and lost time.
Built a unified automation system connecting Notion, email, APIs, and task flows into a single operational layer.
- Clear operational visibility
- Reduced tool switching
- Faster execution cycles
Micro-SaaS: Workflow Builder
Reusable workflow engine designed to turn repeated automation setups into a product.
Same automation patterns were being rebuilt across projects.
Abstracted workflows into reusable modules and built a product-ready backend system.
- Reduced rebuild time
- Foundation for SaaS product
- Reusable automation components
E-commerce Automation Engine
Backend automation system for managing orders, inventory sync, and customer communication.
Manual order processing and tracking caused delays and errors.
Built event-driven automation handling order lifecycle, notifications, and integrations.
- Faster order processing
- Reduced errors
- Improved customer experience
AI Agent for Inbox & Task Handling
An AI agent we built for our own inbox: it triages every email, drafts replies from past answers, sends action items to a task app, and waits for approval before sending.
The costly part of our inbox was the first pass: reading every email to work out what it was, what it needed and when, then writing replies we had written many times before.
An agent that classifies and summarises each email, drafts replies from the most similar past answers, exports action items to Google Tasks through n8n, and holds every reply for approval. Version 2 adds a web interface, four model providers and a public demo.
- Every email arrives sorted, summarised and with its action items listed
- Replies drafted in our own voice, with the past replies they drew on shown
- Nothing is sent without a person approving it
- Open-sourced under MIT, with a demo that needs no account or API key
B2B Lead Generation Pipeline
End-to-end extraction and enrichment machine that creates hyper-personalized outreach sequences.
Sales teams were spending 60% of their day manually researching leads.
Engineered an autonomous agent that scrapes target companies, enriches data, and drafts emails.
- 4x increase in outbound volume
- Eliminated manual SDR research
- Higher reply rates
Customer Support AI Triage
Autonomous resolution engine capable of handling L1 tier support tickets without human agents.
Support queues were bogged down by repetitive password resets and billing inquiries.
Deployed a specialized LLM agent synced with internal knowledge bases to auto-resolve simple issues.
- 40% of tickets auto-resolved
- Instant response times
- Human agents focused on complex issues
Consolidated KPI Reporting Bot
Data pipeline that synchronizes cross-platform analytics into unified Slack notifications.
Founders were logging into 5 different tools daily just to check core metrics.
Built a chron-scheduled worker that polls Stripe, Analytics, and SQL databases, delivering daily summaries.
- Absolute metric transparency
- Zero daily dashboard logins
- Immediate anomaly detection
Zero-Downtime Migration Engine
Scripting and server architecture designed to safely port millions of rows of legacy string data.
Aging MongoDB cluster was failing under load, needing conversion into relational models.
Wrote a batched streaming utility that mapped NoSQL documents onto normalized Postgres tables live.
- 0 seconds of application downtime
- Data integrity preserved
- 95% faster query speeds
The Pattern Behind Everything
What we don't show here
Ready to architect your next system?
If something in your workflow keeps happening again and again, that's a system waiting to be built.