Your AI strategy will fail if your data foundation is weak.
I am the part-time Head of Data companies bring in when AI and reporting stall on a weak data foundation. I set the plan and lead the engineers who deliver it.
Previously at Playtika and EPAM · Warsaw, Poland · Remote worldwide
You do not have a data strategy.
You have data debt.
The model is not the bottleneck.
The cloud provider is not the bottleneck.
The dashboard tool is not the bottleneck.
The real bottleneck is underneath:
- Unclear ownership
- Fragile pipelines
- Slow delivery cycles
- Weak governance
- Platforms never designed for scale
Does this sound familiar?
If several of these feel familiar, the problem is not your team. Bad data leads to bad AI.
Every company has a queue.
I have taken one apart.
At Playtika I helped lead the data platform the whole company ran on, and every new data source had to queue behind one central engineering team that everything and everyone waited on.
I set the direction for the redesign that took that queue apart, and I led the engineers who delivered it, moving ownership of the pipelines out to the teams that actually used the data. They stopped waiting on us.
Before that I led delivery on enterprise transformation at EPAM for the London Stock Exchange, and since March I have published eleven articles on data platforms and AI readiness.
Company size is rarely the bottleneck. It is who everyone has to ask before anything moves.
Seven areas where execution breaks down.
Each one is a place where misalignment between data, AI, and engineering creates the most expensive gaps.
Data Platform Architecture
Design scalable, reliable data infrastructure that grows with the business — from ingestion frameworks to serving layers.
AI Readiness Advisory
Assess what actually needs to happen before AI delivers value. Close the gap between ambition and foundation.
Data Strategy & Governance
Define ownership, trust levels, and quality frameworks. Turn governance from bureaucracy into a speed advantage.
Engineering Leadership Advisory
Strengthen technical direction and architecture decision-making across engineering organizations.
Platform Modernization
Migrate legacy systems and replace fragile pipelines without disrupting operations or losing the team.
Fractional Head of Data
Senior data platform leadership part-time. 1–2 days per week embedded. All the direction. None of the overhead.
Executive Technology Advisory
Translate technical reality into board-level clarity. Align engineering decisions with business outcomes.
The IT Nomad Lab Method
Four phases. Applied consistently to every engagement, at every scale.
Diagnose the Foundation
Map data flows, identify fragile components, and understand where trust started to break. The first step is understanding how the system actually works — before changing anything.
→ System map + ranked blockers, week 2
Design for Scale
Translate diagnosis into architecture that supports growth and execution. Identify quick wins and sequence the foundation work that must happen before AI ambitions become realistic.
→ 90-day stabilization plan, week 5
Enable the Teams
Reduce dependency loops. Move ownership closer to builders. Create patterns that allow teams to add new data sources without creating more chaos.
→ Self-service enablement framework
Deliver Measurable Impact
Turn architecture into adoption, execution, and business outcomes. Exec-ready roadmap covering architecture, sequencing, team gaps, and budget.
→ AI-readiness roadmap + exec deck, week 8
Understand the system → Stabilize the foundation → Enable the teams → Deliver measurable impact.
Josias De Lima
Founder, IT Nomad Lab
Engineering leader. Data platform architect. Execution-focused technology strategist. I work at the intersection where data infrastructure, team operating models, and business outcomes either connect or fail.
My background spans ingestion frameworks, ETL orchestration, data lake architecture, governance programs, self-service enablement, and the organizational design that makes platforms actually get used. I have led this work at Playtika, on enterprise transformation at EPAM for the London Stock Exchange, and in fast-moving product companies where speed and reliability both had to be true at the same time.
The pattern I keep seeing: organizations invest heavily in technology and move slowly anyway. Not because the technology is wrong. Because the architecture, the team operating model, and the ownership structure were never designed to scale together.
Complex problems do not require magic. They require clarity, persistence, and consistent execution.
Writing on data, AI, and engineering execution.
Published every few weeks. For CTOs, Heads of Data, and engineering leaders.
Questions worth asking.
Direct answers to the questions that come up in every discovery call.
A data platform consultant helps organizations design, build, and optimize the infrastructure that stores, processes, and delivers data across the business. This includes architecture decisions, pipeline design, governance frameworks, and team enablement strategies — with the goal of making data reliable, fast, and owned by the teams that use it.
Before you scale AI, fix the foundation.
If your teams are moving fast but your data still feels slow, fragile, or unclear — we should talk. A 30-minute discovery call is the fastest way to know if what I do maps to what you need.
No pitch deck. No follow-up sales sequence. Just a structured conversation.