Consulting - Rob Winters

I work with a small number of early-stage and scaling companies who need hands-on data engineering and AI expertise—but aren't ready for (or don't need) a full-time hire. 10-20 hours a week, embedded in your team, focused on real delivery.

Services

Data Engineering & Architecture

Pipeline design, warehouse architecture, data quality systems. Hands-on implementation with dbt, Snowflake, Airbyte, and the modern data stack.

AI & LLM Integration

RAG systems, AI agents, LLM workflows. Practical AI that fits your existing data infrastructure—not hype-driven projects that go nowhere.

Infrastructure & DevOps

Infrastructure as code, observability, CI/CD for data systems. Engineering practices that scale from prototype to production.

Advisory & Problem Diagnosis

Architecture reviews, technical roadmaps, troubleshooting. A second opinion when you're stuck or need an outside perspective.

How I Work

Hands-on, not hands-off: I work with your team to write code, build pipelines, and implement solutions. You get a practitioner, not a slide-deck consultant. I think the best way to understand a problem is to work on it.
Capacity building: The best outcome is when your team can run things without me. I document everything, transfer knowledge, and set you up for independence. If I haven't made myself redundant, I haven't done the job.
Selective by design: I take on 2-3 clients at a time. This isn't a funnel—it's how I ensure every engagement gets real attention. I'd rather do fewer things well than many things poorly.
Honest about fit: If I'm not the right person for your problem, I'll tell you. Sometimes the most valuable thing I can offer is a referral to someone better suited. No point wasting either of our time.

Engagement Models

Model	What It Looks Like	Typical Duration
Fractional	Ongoing advisory + hands-on implementation. Embedded in your team, regular cadence.	Monthly retainer
Architecture Review	Deep-dive assessment of your current setup, written recommendations, technical roadmap.	2-3 weeks
Sprint Project	Specific build or implementation with clear scope. Pipeline, integration, or proof-of-concept.	2-6 weeks

Tech Stack

Data

dbt, Snowflake, BigQuery, Airbyte, Meltano, Tableau

Infrastructure

Terraform/OpenTofu, Docker, GitHub Actions, AWS

AI & LLM

OpenAI, Anthropic, self-hosted inference, RAG systems

Languages

Python, SQL, TypeScript

Why Work With Me

20 years experience, still hands-on: I've led data at scale, but I still write code daily—dbt since 2019, Snowflake since 2019, Tableau since 2011. I think you can't lead effectively if you've lost touch with the craft.
Built teams at Miro and Navan: I know what good looks like at scale. I've built 30-person global data teams and the platforms they run on. But I'm happier building and coaching than managing these days.
AI/LLM practitioner: I'm building real systems with LLMs—not theorizing about them. RAG pipelines, agents, production deployments. Most AI work is hype; I'm interested in what actually works.
Async-friendly, globally tested: Based in France, I've spent years working with distributed teams across time zones. Async is native to how I work. I don't need meetings to feel productive.

Recent Work

KNVB (Royal Netherlands Football Association)

Designed and built a new data platform from the ground up over an 18-month engagement. Legacy infrastructure was replaced with a modern stack (Terraform, Azure, Dagster, dbt, Tableau) that supports the organization's analytics needs across competitions, membership, and operations.

Next Steps

The best way to start is a conversation. Tell me about your challenge—what's working, what's not, what you're trying to build. I'll be honest about whether I can help, and if I'm not the right fit, I'll tell you straight. No point wasting time.

I typically respond within 24-48 hours. If it seems like a fit, we'll schedule a call to dig into the details.