🎯 Mark Freeman II

onthemarkdata.com | Senior data engineer helping brands with GTM and content strategy.

I'm a community health advocate turned data engineer interested in the intersection of social impact, business, and technology. My life’s mission is to improve the well-being of as many people as possible through data—especially among those marginalized.
 
As the founder of On the Mark Data, I use this platform to help brands connect to data professionals through captivating content.
 
I received my M.S. from the Stanford School of Medicine and I'm also certified in Entrepreneurship and Innovation from the Stanford Graduate School of Business. In addition, I have worked within numerous startups where I have put machine learning models into production, integrated data analytics into products, and led migrations to improve data infrastructure.
 
The culmination of my technical skills, content success, and experience within startups would make me an asset whether you are looking to create a successful content strategy, engage in influencer marketing campaigns, or develop a go-to-market strategy for your startup.

You can learn more about opportunities to collaborate, my media appearances, or my content at onthemarkdata.com.

Tech Lead, GTM Engineering

• Joined as employee one to lead GTM at Gable to help us reach our first customers.
• Worked with the CEO to establish our GTM strategy, content strategy, and messaging for technical audiences.
• Building internal data products to drive our GTM strategy and sales.

Hannes Mühleisen (co-founder of DuckDB) doing his slides like SQL queries is so damn cool.

The most powerful communication is often what's not said.

Dev 🤝🏼 GTM | O’Reilly Author | [in]structor (33k+ Learners)

No alternative text description for this image

Joe Reis recently inspired me to consider Claude Code as a writing tool when he challenged the notion of context graphs.

He essentially took the Foundation Capital article that has been going viral and ran it through Claude Code.

It instantly transformed a very theoretical article into an entire code repo you could explore on his public GitHub (then used that as a tool to critique).

A thought then popped into my head:

"What if my writing was so clear that you could throw an entire article into Claude Code and it would one shot a working POC of the presented ideas."

Welp, the implementation chapter of our book (Ch. 7) is public on GitHub and is stored as a README file.

So let's see if my writing is clear enough for AI to one shot it.

I downloaded the repo to my local machine, got rid of every file except the README, the JSON dataset, and a folder of diagrams from the chapter.

Then I prompted: "This is my technical chapter of the book. Implement it."

I couldn't believe it... In about 10 minutes it was spinning up a Docker image, loading data into Postgres, and then running my tests.

Keep in mind I spent a couple intense weeks building that codebase for the book.

To be clear, I'm not advocating for such code to be useable, but rather it's now a tool that I can use to further validate my writing.

Present your novel ideas.
But show the working POC.

It's the same as in academia where you have research papers that include code and data for repeatability.

The bar has risen for everyone now. Talk is cheap.

Crazy Idea: I use claude code on a live stream where I'm building an app from scratch.

But...

People on the live stream can enter prompts on my behalf while I'm building to either help or mess with the project.

I think the chaos would be both fun and hilarious.

You didn't hear it from me, but the coolest talk at Data Day Texas is happening at 2:55PM in room 204 👀

Live coding exercise for my conference talk at Data Day Texas? You bet!

Saturday - 2nd Floor - Ballroom 204

Code: The Untapped Metadata Source Driving Most Data Failures

Over the past three years, collaborating with organizations implementing data contracts and writing the O'Reilly book on the topic, it has become abundantly clear that data is surprisingly the wrong surface to address systemic data issues —we actually need to focus on code. Code not only defines what data is captured and moved through software systems, but it's also the ultimate source of truth in how a business operates and its documented changes over time. This is evident in how some of the most significant internet outages can be traced not only to code changes, but also to how those changes affected underlying data models. In this talk, I share case studies and lessons learned in extracting metadata from code and using it to enforce expectations across software systems and data products.

Remember back in the ancient internet times when you could use this cool thing called HTML to customize your Myspace profile.

Just randomly hacking away at sections to see what changed. Copy and pasting random code and hoping it works.

Vibe coding feels eerily similar.

What data scientist experienced in 2023 with the realease of ChatGPT is the same experience software engineers are experiencing today with Claude Code.

Back in 2022 I was transitioning from data science to data engineering, and a major project that led to that promotion was building and deploying a natural language processing pipeline.

Back then I was using SpaCy, which was an opinionated framework for implementing the latest production-ready NLP models.

Yeah... I open up ChatGPT and immediately see the "advanced" work I put into production was now antiquated.

While I wouldn't throw an LLM at my production project back in 2023, I was intimately aware of what was required to get business value out of text.

What ChatGPT showed me was that a large portion of the difficult parts of NLP were now just a simple API call... and more importantly, highlighted the inflection point if the momentum of change followed.

Fast forward to today, and so many software engineers built side projects over the holiday break and experienced the exact realization I had back in 2023.

"This LLM accomplished way more than I expected, and I can see an inflection point if this trend continues."

Imagine what will be accomplished in the next few years.

I help entry-level, non-technical professionals get their first data analyst role | Data Analyst @RBC | Data Career Coach

I help immigrants and career-switchers land analyst roles—without burning out on certificates.

By day I build decision-grade analytics (currently at Autodesk). By night I run Data Career Compass, where we treat your job search like a product launch:
Market → Positioning → Portfolio → Distribution → Interviews.

My POV:
Most candidates try to add more tools. Winners learn to package their value. Your portfolio isn’t a gallery; it’s a sales page for your judgment.

Receipts:

→ Professionals land interviews by aligning projects to business outcomes (not toy datasets).

→ My frameworks have been used across finance, tech, and education to translate experience into results.

→ Built data products and controls in regulated environments (RBC Capital Markets) that leaders rely on daily.


What you’ll get here:

→ Weekly breakdowns of real projects (SQL, Python, dbt, DuckDB, Looker/Tableau/Power BI).

→ Storytelling patterns that turn insights into action.

→ Templates that remove friction: portfolio briefs, stakeholder one-pagers, interview scorecards.

Coming February:

I’m launching a coaching program designed for mid-level and senior professionals ready to transition into analytics.

This isn’t an entry-level bootcamp—it’s a career accelerator that helps you reposition your existing experience, communicate business value, and step confidently into data-driven roles.

Work with me:

↳ Cohorts + 1:1 coaching → DM “HIRED”

↳ January program for mid/senior professionals → DM “JANUARY” for early access

↳ Guest workshops for teams and communities → DM “WORKSHOP”


I build analysts who think like product managers and communicate like consultants.

Peter Ndiforchu

Most candidates think they’re being evaluated on what they built.

They’re not.

When you explain a project, hiring managers are quietly assessing something much deeper.

Here’s what’s actually happening behind the scenes 👇

1️⃣ Can you think in business outcomes—or just tasks?
If your story starts with “I used SQL to…” and ends there, that’s a red flag.
They’re listening for why the project existed and what changed because of it.

2️⃣ Do you understand the problem—or just the solution?
Strong candidates explain the constraint, the tradeoff, and the decision.
Weak ones jump straight to the dashboard.

3️⃣ Can you simplify complexity?
If you can’t explain your project to a non-technical stakeholder in 60 seconds,
they assume you’ll struggle in real meetings.

4️⃣ Did you drive the work—or were you just assigned it?
Ownership shows up in language:

“I noticed…”

“I proposed…”

“I pushed back when…”


That’s leadership without the title.

5️⃣ Do you sound like someone they can put in front of executives?
Clarity. Structure. Calm confidence.
That’s what earns trust.

Here’s the uncomfortable truth:

Two candidates can build the same project.
Only one gets the offer—because they frame it differently.

Your projects don’t get you hired.
Your interpretation of them does.

If this resonated, next time you explain a project, ask yourself:
“Am I telling a story—or reciting a task list?”

Helping Non-Technical Professionals Build Data-Driven Careers That Travel Globally | Data Career Strategist | Senior Data Analyst @ Autodesk

What Data Hiring Managers Really Evaluate on Interviews

Most candidates know the STAR method.
Very few use it in a way that actually signals seniority in data roles.

Here’s how hiring managers experience it 👇

The STAR method for data roles (with a real example)

S — Situation
❌ Weak: “We didn’t have good dashboards.”
✅ Strong: “Sales leaders were making weekly decisions using spreadsheets pulled manually from three systems.”

Why this matters:
You’re showing business context, not just a messy data problem.

T — Task
❌ Weak: “I was asked to build a dashboard.”
✅ Strong: “I was responsible for creating a single source of truth that leadership could trust for forecasting.”

Why this matters:
This reframes you from executor to owner.

A — Action
❌ Weak: “I used SQL and Tableau.”
✅ Strong:
“I audited existing tables, wrote validation queries to reconcile numbers across sources, aligned definitions with Sales Ops, and designed a dashboard focused on decision-making—not vanity metrics.”

Why this matters:
Tools are expected. Judgment is not.

R — Result
❌ Weak: “They liked the dashboard.”
✅ Strong:
“Forecast variance dropped by 18%, manual reporting time was reduced by 6 hours per week, and the dashboard became the default reference in leadership meetings.”

Why this matters:
Impact beats effort. Every time.

The real mistake candidates make:
They treat STAR like a storytelling template.
Hiring managers hear it as a decision-making framework.

In data roles, STAR isn’t about what you built.
It’s about:

How you understood the business

How you made trade-offs

How your work changed behavior


If your STAR answers don’t reveal those signals,
you’re underselling yourself—no matter how strong your skills are.

📌 Save this before your next interview.

How I went from Cameroon → Canada → a US data role (without ‘local experience’).

When I moved, I assumed the biggest hurdle would be visas, credentials, or “local experience.”

It wasn’t.

The real question global recruiters were silently asking was:
“Who can vouch for this person when we can’t observe them directly?”

Here’s what made the difference on my journey 👇

1/ Relationships reduce hiring risk
Recruiters outside your country can’t see your work day-to-day. They rely on people they trust to validate your judgment and impact.
I focused on building genuine connections with:

Hiring managers

Senior data practitioners

People in the ecosystem who could confidently say,

 “I’ve worked with him. He delivers. I’d hire him again.”

2/ Visibility beats invisibility
I didn’t just rely on a resume. I made my work and thinking visible by:

Sharing real projects publicly

Discussing trade-offs and decisions

Engaging with communities in LinkedIn, Slack, and Discord

This gave recruiters context before even meeting me.

3/ Skills matter, but signals win
SQL, Python, Excel—all necessary. But what got me hired was showing:

Structured thinking

Clear communication of impact

Ability to navigate ambiguity

Coupled with endorsements from people who could vouch for me, geography became irrelevant.

Global hiring isn’t about ‘local experience’—it’s about trusted signals.

Ask yourself: “Who in this market can confidently vouch for my work?”

That’s the real game.

Cameroon → Canada → a US data role

The first time I realized hiring expectations weren’t universal…
I thought I had done everything right.

Strong technical skills.
Solid projects.
Clear results.

Then I walked into an interview outside my home market and something felt off.

They weren’t impressed by what tools I knew.
They kept probing how I worked.

Questions like:

“How did you influence stakeholders who didn’t agree with you?”

“What trade-offs did you make under pressure?”

“Who trusted your judgment and why?”


That’s when it clicked.

In some markets, hiring is about capability.
In others, it’s about credibility.
And in global roles, it’s about signal alignment.

Same resume.
Same experience.
Very different evaluation lens.

I stopped optimizing for: 
❌ tool checklists
❌ keyword stuffing
❌ “local experience” myths

And started optimizing for: 
✔ decision-making stories
✔ business judgment
✔ proof that someone would trust me in the room

That shift changed everything.

Global hiring isn’t harder.
It’s just calibrated differently.

If you’re preparing for roles beyond your home country, ask yourself: Are you proving skills…
or earning confidence?

Because globally, confidence hires before competence.

Here’s the uncomfortable truth most candidates don’t hear early enough:

Most data portfolios aren’t evaluated as “work.”
They’re evaluated as signals.

And most of them send the wrong ones.

I’ve reviewed a good number of portfolios as part of the hiring team.
The pattern is consistent:

What candidates show:
• Dashboards with no business question
• Kaggle-style analyses disconnected from real decisions
• Tools-heavy walkthroughs (“I used SQL, Python, Tableau…”)
• Perfect charts that solve imaginary problems

What hiring managers are actually scanning for:
• Can this person frame a messy business problem?
• Do they understand tradeoffs, not just techniques?
• Can they explain why something mattered — not just how they built it?
• Would I trust them to influence a decision with imperfect data?

Here’s the part most people miss:

A portfolio isn’t proof of skill.
It’s evidence of judgment.


That’s why two candidates can use the same dataset —
and only one gets a callback.

The one who: 
• Anchors the project to a real stakeholder decision
• Explains assumptions and constraints
• Shows what they didn’t do and why
• Communicates impact in plain business language

Strategic shift to make today:

Stop asking, “How do I show more tools?”
Start asking, “What decision does this help someone make?”

That question alone will put your portfolio ahead of 90% of applicants.

If you’re building a portfolio right now, build it like you already have the job — not like you’re trying to pass a course.

taplio

The Battle

🎯 Mark Freeman II

Peter Ndiforchu

Followers

Posts

Engagements

Likes

Performance Score🎯 Mark Freeman II's LinkedIn Performance

Performance ScorePeter Ndiforchu's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

🎯 Mark Freeman II

Peter Ndiforchu

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}🎯 Mark Freeman II's LinkedIn Performance

Performance ScorePeter Ndiforchu's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance Score🎯 Mark Freeman II's LinkedIn Performance