Digital Systems Architect · Data Scientist
I design the systems and data your business runs on.
I work with businesses that have outgrown improvised tools — designing the architecture that connects their systems, the pipelines that move their data, and the models and analytics that turn both into decisions.
Systems that stay coherent as they grow, and data you can actually decide with — designed for real operations, not hype.
Who I work with
Businesses whose systems and data have grown faster than the structure holding them together.
Founders building the foundation
Owners who want the system architecture and data model designed correctly, before it becomes expensive to change.
Teams drowning in disconnected tools
Companies running on tools that don't talk to each other, with data scattered across spreadsheets, CRMs, and inboxes.
Operations that need to measure
Businesses that have data but no reliable way to model it, trust it, or turn it into decisions.
What I do
Two connected disciplines: the architecture that holds a business's systems together, and the data layer that makes them measurable.
Systems Architecture
Systems & Integration Architecture
Designing how a business's systems fit together — data flow between tools, integration and API design, ownership boundaries, and an architecture that stays coherent as the company grows.
CRM & Operating Model Design
Designing CRM structure, data model, pipelines, and workflow logic around how the business actually operates — so the system reflects reality instead of fighting it.
Automation & AI Agent Systems
Designing automation and AI-assisted workflows for operations, lead handling, support, and internal processes — with explicit business rules and logic behind them.
Security, Identity & Infrastructure
Access architecture, identity and 2FA policy, device and data governance, network design, and Google Workspace and email infrastructure (SPF, DKIM, DMARC).
Data & Analytics
Data Architecture & Pipelines
Data modeling, ETL and pipeline design, SQL, and warehouse structure — building one reliable source of truth instead of numbers that disagree across tools.
Predictive Modeling & Machine Learning
Python-based models for forecasting, churn and retention, lead scoring, and segmentation — scoped to problems where a model genuinely beats a rule.
Analytics, KPIs & Decision Systems
KPI definition, dashboards, and reporting built so the people making decisions can actually read them — Looker Studio, Power BI, Metabase, and the metric definitions underneath.
Measurement, Tracking & Attribution
Event tracking, conversion and funnel instrumentation, and attribution design — so marketing, sales, and product activity can be measured against real business outcomes.
Topics I cover
Practical writing on systems architecture, data infrastructure, and making decisions from data.
Designing Systems Before They Become Legacy
Architecture decisions that are cheap now and expensive in two years — and how to tell them apart early.
Building One Source of Truth
Why numbers disagree across tools, and how data modeling and pipeline design fix it at the root.
From Data to Decisions
KPI definition, dashboards people actually read, and the difference between reporting and insight.
Predictive Modeling in Small Businesses
Where forecasting, churn, and lead scoring genuinely pay off — and where a simple rule beats a model.
CRM as a Data Model, Not a Tool
Designing CRM structure, fields, and pipelines as a data model that supports both operations and analysis.
AI Agents & Practical AI for Business
Where AI agents and AI-assisted workflows fit into support, operations, and internal processes without falling for hype.
Measurement, Tracking & Attribution
Instrumenting events, funnels, and conversions so spend and effort can be tied to real outcomes.
Access, Identity & Data Governance
Practical security architecture: who can reach what, on which device, and who owns the data.
Network & Infrastructure Basics
Internal network design, segmentation, and reliability for small and medium businesses.
Hiring & Working with Technical People
How founders should hire, brief, and work with engineers, analysts, and IT teams.
Technical Decisions & Architecture Strategy
Choosing tools, architectures, and priorities before mistakes start to compound.
FAQ
Common questions from founders and teams building systems and data infrastructure.
What does a digital systems architect actually do?
I design how a business's systems, data, and workflows fit together — the structure underneath the tools. Tools get replaced eventually; the architecture decides whether replacing one takes a weekend or a year.
How is this different from hiring a developer or an IT provider?
A developer builds what they are asked to build, and an IT provider keeps things running. I decide what should be built and how the pieces should connect — the design decisions that come before implementation and outlast it.
Do we have enough data for data science?
Usually more than you think, and usually messier than you think. Most businesses need data architecture and reliable measurement first; modeling only pays off once the underlying numbers can be trusted.
When is a predictive model worth it, and when is a rule enough?
A model earns its place when the pattern is real, changes over time, and the decision repeats often enough to matter — forecasting, churn, lead scoring. If a simple rule captures most of the value, I will tell you to use the rule.
Why do our numbers never match between tools?
Because each tool defines the metric its own way and nothing reconciles them. The fix is a defined data model and a pipeline that produces one source of truth, not another dashboard on top of the mess.
Do small and medium businesses really need a CRM?
Most need a clear data model for leads, follow-up, and customers much earlier than they think. A properly designed CRM is an operating system for the business, not just a sales tool.
What does AI agent design mean in practice?
Designing AI-assisted systems for tasks such as lead qualification, support routing, internal guidance, and repetitive operational work — with explicit business rules, boundaries, and failure handling behind them.
Can AI replace technical judgment?
No. AI accelerates work, and it accelerates mistakes just as efficiently when the fundamentals are missing. It is a tool, not a substitute for understanding the system.
What should we fix before scaling or spending more on marketing?
A reliable data and measurement layer, a clear CRM structure, sane access control, and a follow-up process that actually runs. Growth amplifies whatever weakness is already there.
Do you work with startups or with established companies?
Both — early-stage businesses designing the foundation, and established small to medium companies whose systems and data have outgrown the structure holding them together.
Which tools and technologies do you work with?
Python and SQL for data work, standard pipeline and warehouse tooling, BI tools such as Looker Studio, Power BI, and Metabase, and the CRM, automation, and Google Workspace stack most businesses already run on.
How does an engagement usually start?
With a short review of your current systems, your data, and the decisions you are trying to make. That is usually enough to say what should be fixed first, what can wait, and whether I am the right person for it.
Contact
If your systems have outgrown their structure, or you have data you cannot yet make decisions with, tell me where things stand and I will tell you what I would fix first.
To get a fast, useful answer, include:
- Your business name and website (or domain)
- Your team size
- What you want to improve: architecture, CRM, automation, data infrastructure, or analytics
- Your current stack (CRM, tools, databases, dashboards, email, automation)
Quick note:
Most operational problems are architecture problems wearing a different costume: disconnected tools, no clear data model, metrics nobody agrees on, and no technical ownership. Fixing the structure is what makes everything above it work.
Design the system. Trust the data.