Services

What I build, and why it actually matters for your business

Here is the thing about automation and data work: nobody wakes up excited about ETL pipelines or webhook architecture. What people care about is the result. Fewer hours lost to manual work. Leads that do not slip through the cracks. Reports that show up on time without someone staying late to build them.

So here is what I build, organized by the actual problem each one solves.

Data pipelines and ETL

The problem

Your data lives in five different places, and getting it into one usable format means someone exporting CSVs, copy-pasting into spreadsheets, and praying nothing breaks.

What I build

Reliable pipelines that pull data from wherever it lives, whether that is Bloomberg feeds, SQL databases, SaaS APIs, or shared network drives, clean it up, and land it exactly where it needs to be. No more manual exports. No more "wait, is this number even current."

In practice

At OMERS, I built pipelines handling live financial data from Bloomberg SAPI and MSCI APIs into PostgreSQL, used daily by risk and portfolio management teams.

Have a data mess you have been avoiding? Tell me what's going on and I will tell you what it would take to fix it.

Read the full guide: Data pipelines & ETL, explained without the jargon →

More guides: ETL vs ELT · Real-time pipelines · Airflow vs Prefect · Data quality & observability

AI lead qualification and chatbots

The problem

Leads come in faster than your team can manually review them, and by the time someone gets to a hot lead, it has gone cold.

What I build

AI systems using OpenAI and Claude that read incoming leads, score them against your specific criteria, and route them automatically to the right person. Chatbots that handle real conversations, not just canned FAQ responses, and know when to bring in a human.

In practice

For a marketing agency client, I built an end-to-end system that captures leads, scores them with AI, nurtures the ones that are not ready yet, and books appointments automatically once a lead qualifies. The team only handles the conversations that actually need them.

Curious what AI-driven lead qualification could do for your pipeline? Let's talk through it.

Read the full guide: AI lead qualification, how to stop losing hot leads →

More guides: AI agents for business · AI chatbots · RAG on your data · Support automation · Appointment booking

Workflow and business automation

The problem

Someone on your team spends hours every week doing the same repetitive task: pulling reports, updating spreadsheets, sending the same follow-up emails.

What I build

Automations using n8n and Python that handle the repetitive work permanently. Triggers based on events, schedules, or incoming data, with proper error handling so the automation does not silently fail at the worst possible time.

In practice

One client's team was spending over 25 hours a week manually pulling campaign data and writing client reports. The automated version runs overnight and lands in the client's inbox every morning, no human required.

What's eating your team's time right now? Describe it here and I will tell you honestly if automation is the answer.

Read the full guide: Workflow automation with n8n and Python →

More guides: n8n vs Zapier vs Make · n8n development · Business process automation · CRM automation · Document processing

Cloud infrastructure and dashboards

The problem

Your systems either do not scale, or they scale and the bill becomes a monthly surprise.

What I build

Cloud architecture on AWS, GCP, or Azure designed for your actual stage and budget, whether that is serverless functions for a lean startup or containerized infrastructure for something more established. Plus dashboards in Streamlit or React that actually get used, because they are built for the people looking at them daily, not for a one-time demo.

Wondering if your current setup is costing you more than it should? Ask me directly.

Read the full guide: Cloud infrastructure and dashboards that get used →

More guides: AWS vs GCP vs Azure · Streamlit dashboards · Serverless for startups