You need better metrics!

Your operation is losing margin to something you can feel but cannot measure.

I spent a decade at CERN finding the one variable that mattered among millions. Now I find it on your operations floor — the structural friction quietly capping your margin before anyone tries to sell you a solution.

From CERN to the operations floor

What makes the physics relevant isn't the theory. It's a decade of practice in places where being wrong meant losing years of work.

For more than ten years at CERN, I built the control systems behind some of the most demanding physics experiments ever run — in environments where a single unmeasured variable could invalidate years of work.

The discipline that requires is simple to name and very hard to live: know which variable actually matters, and measure it before you act.

That discipline is what I now bring to deep-tech and engineering-driven organizations across Europe, including precision manufacturers, robotics companies, semiconductor firms, and industrial automation providers. My path — experimental particle physics in Germany, then quantum computing hardware in Finland alongside teams from IBM, Intel, and Google, then deep-tech strategy in Helsinki — is unusual on purpose. I speak to your engineers peer to peer, I understand your systems at a physical level, and I translate both into a decision you can act on the same week.

10+ yrs

Building high-stakes control systems at CERN

3

Tech giants — IBM, Intel, Google — among the end customers of my quantum-hardware work

1

Point of accountability across every implementation partner

4–5

Network-native implementation partners across cloud, data, and industrial AI — each 5+ years operating, 20+ engineers, selected for proven delivery in European deep-tech environments

5

Recurring patterns that silently cap yield — named before any tool is bought

// margin recovery — opportunities the diagnostic typically surfaces

Ex-CERNMSc / PhD Experimental Particle PhysicsJohns Hopkins AI Certified (Applied Generative & Agentic AI)QSilver Certified (Quantum Computing & Algorithms)Electronics EngineeringLean Six SigmaISO Standards

The patterns your data is hiding

Something is limiting your operation. Your instinct is right. Your metrics just aren't catching it yet.

You're running a technically capable operation. Your team is good. Your equipment is adequate. And still — throughput isn't where it should be, and margins are tighter than the numbers explain. You sense a structural friction. But every time you push for the data to prove it, the picture fragments. That isn't an AI problem. It's a systems-visibility problem. Five patterns show up again and again in deep-tech and engineering-driven organizations across Europe:

P01

Your data exists but doesn't speak.

Production, quality, and ERP records are all there — and all siloed. The complete operational picture is never in one room at one time.

P02

The bottleneck is sensed, not measured.

Your operations lead almost always knows where throughput is lost. But intuition isn't a metric, and without a quantified constraint, improvement efforts scatter.

P03

Critical knowledge lives in people, not systems.

When the right individual leaves, the knowledge leaves with them. What looks like institutional expertise is a single point of failure.

P04

Quality failures are absorbed, not solved.

Rework, scrap, and late-stage defects quietly eat margin, because their true cost is rarely calculated and is treated as the cost of doing business.

P05

Technology was added without integration.

ERP, MES, automation — each solved one problem. Without a layer connecting them, you get more systems and the same blind spots.

Recognise more than two? Then the friction is structural. Which means it can be quantified — and resolved.

Why this holds up

I'm early in this practice. I'm not early in the work it's built on.

I won't show you testimonials I haven't earned yet. JMP Ventures is new, and the engagements running today are proof-of-concept pilots, not finished case studies. I'd rather tell you that plainly than dress up thin proof — that's exactly the discipline I'd bring to your operation. So here is what the work actually rests on:

Pedigree and ecosystem are the proof until clients are. Honest, and stronger than a borrowed quote.

01A decade inside the world's most unforgiving measurement environment.

The diagnostic rigour isn't borrowed from a framework. It was the daily standard at CERN, where imprecision wasn't an option.

02A network-native implementation ecosystem with the track record I'm still building.

I don't execute alone. I orchestrate 4–5 established European partners across cloud, data privacy, and on-premises AI — each operating 5+ years with 20+ engineers. You inherit their proven delivery; you manage none of it.

03One point of accountability across all of it.

Whatever the partner mix, you talk to me. The strategy and the execution answer to the same person.

How we'd work together

Four ways to engage, depending on where your operation is today.

Each one starts with listening. None starts with a pre-decided technology.

Abstract render: a single constraint located within a field of cubes

ENGAGEMENT 01

The Operational Stress-Test

For deep-tech and engineering-driven organizations that suspect structural friction but haven't quantified it. A structured diagnostic sprint across your production, quality, and decision flows. You leave with a working map of the constraints costing you margin — not a report that sits on a shelf.

  • Know exactly where your yield is being lost
  • A prioritised list of leverage points, ranked by impact
  • A clean, evidence-based basis for any technology decision that follows

Right for: Managing Directors and founders who want evidence before investment.

Abstract render: a layered architectural tower of translucent system strata

ENGAGEMENT 02

Systems Mapping & Industrial AI Strategy

For operations ready to move from diagnosis to roadmap. I build the strategic architecture — how AI, data systems, and process redesign fit into one coherent, scalable operating model. The partners who'll implement are already selected and briefed.

  • A vendor-neutral strategy built around your existing infrastructure
  • No rip-and-replace disruption
  • A roadmap your team can follow and your investors can understand

Right for: Companies modernising without losing what already works.

Abstract render: a central orchestration hub coordinating orbiting nodes

ENGAGEMENT 03

Strategic Orchestration Retainer

For companies in active implementation who need one point of accountability. I manage the partner ecosystem — AI providers, data specialists, cloud, cybersecurity — across the whole implementation. You talk to one person.

  • Implementation stays on strategy, not on whoever shows up
  • Partner performance managed proactively
  • Yield improvements tracked against the original diagnostic

Right for: Operations that can't afford for implementation to drift from strategy.

Abstract render: an expansive interconnected system of cubes and data columns

ENGAGEMENT 04

Full Ecosystem Transformation

For organizations committed to becoming systems-dependent rather than individual-dependent. The complete programme: diagnostic, strategy, partner orchestration, and a structured transition to a scalable, AI-integrated operating model competitors can't reverse-engineer.

  • Comprehensive re-engineering of your operational intelligence
  • Operations that run independently of any single expert
  • A systems architecture positioned for the next wave of industrial computing

Right for: Founders with a long-term view of operational sovereignty.

The Physics for Business™ Method

Most consulting recommends new tools. This finds why your current system underperforms — then intervenes where the leverage is highest.

At CERN, no experiment begins with a purchase. It begins with a question: what are the real variables, and how do they interact? The same logic runs your operation.

STEP 1

Identify the real variables

Before any solution is proposed, I map what's actually driving performance — not what the dashboard shows, but what the data beneath it reveals. Production rates, quality events, decision latency, knowledge concentration. The variables that move your margin.

STEP 2

Map the dynamic interactions

Individual metrics lie. The relationships between them tell the truth. I trace how your production, quality, and operational data influence one another — the feedback loops and dependencies that surface as unexplained throughput loss.

STEP 3

Locate the core constraint

Every complex system has one binding constraint that limits everything downstream. Finding it precisely is the difference between a scattershot improvement programme and one that compounds.

STEP 4

Intervene where the leverage is highest

With the constraint located and the interactions mapped, the intervention gets precise — the smallest change that produces the largest recovery in margin. Not a platform migration. Not a broad AI rollout.

The engineer behind the method

The gap between brilliant engineering and operational reality is exactly where deep-tech organizations lose money.

Jorge Mercado, founder of JMP Ventures

— Jorge Mercado, Founder
JMP Ventures · Helsinki

I grew up in Mexico, in a house shaped by two opposing disciplines. My mother moved through complexity by instinct and agility. My father, trained in theology and philosophy at the Vatican, taught me that the most dangerous thing in any system is the assumption no one has examined.

I went to Europe to study physics. I stayed to do the work.

For over a decade at CERN, I built control systems for particle-physics experiments where one unmeasured variable could invalidate years of effort. The job wasn't only precision. It was knowing which variable actually mattered.

Then I moved to Finland and into quantum computing, helping teams at companies like IBM and Intel understand where emerging architectures were genuinely useful. And across industry after industry, I kept seeing the same thing: technically sophisticated organisations drowning in fragmented data and uncoordinated technology — a wide gap between what their systems could theoretically do and what they were actually producing.

The consulting offered in response was either generalist or vendor-captured. Neither started with a diagnosis.

So I built JMP Ventures to be the thing that was missing: one point of strategic accountability, applying the diagnostic rigour of complex-systems physics to the operational problems of deep-tech and engineering-driven organizations across Europe. Higher throughput. Better yield. Decisions that rest on systems, not on whoever happens to still be in the building.

My clients are usually engineers themselves — technically fluent founders and managing directors who don't need anything simplified. They need someone who can sit across the table, speak their language precisely, and hand them a number instead of a narrative.

What this work is for

The goal isn't efficiency. It's operational sovereignty.

An operation that performs independently of any single expert, vendor, or tool. Systems that outlast the people who built them. Decisions grounded in data, not institutional memory.

Deep-tech and engineering-driven organizations across Europe are technically sophisticated and operationally frustrated. They have the technology. They have the people. What's missing is a coherent intelligence layer connecting the two.

The companies that win the next decade won't be the ones that bought the most AI. They'll be the ones that understood their own systems clearly enough to deploy it precisely.

The 5 Patterns Diagnostic

Not sure the friction is structural? Start here.

"The 5 Operational Patterns Quietly Capping Yield in Deep-Tech Environments" — a self-assessment used by industrial leaders across Europe to find where their operations are leaking margin.

Fifteen minutes, and you'll know:

  • Which of the five patterns are live in your operation
  • Where throughput loss is most likely happening
  • Whether the friction you sense is operational, systemic, or structural
  • What a logical next diagnostic step looks like

Stop managing the friction.
Start measuring it.

One structured conversation. One clear map of where your margin is going. Nothing pre-sold before the diagnosis is done.

  • Know within 20 minutes whether a structural constraint is limiting you
  • Leave with a specific area of focus — not a general recommendation
  • Talk to a peer who has worked where imprecision wasn't an option
  • One point of accountability, diagnostic through implementation