Constraint-based planning · Michigan
Most plans are built from averages and last year’s pattern, then collide with reality. Ours start from reality: the actual openings, the courses and credentials that lead to them, the schedules, the places, the people who are available, and the time there is. We build only plans that satisfy all of those limits at once — and when no plan can, we name the limits that bind, so you know what would have to change.
Michigan Workforce Intelligence · Sep 30, 2026
Michigan's August workforce data: what the numbers say, and what they don't
Michigan's August unemployment was 5.0% — but the rate conceals a participation story. The August workforce edition: five decisions for education and industry.
The Axiovex Signal · Nov 12, 2026
How MAST-P Is Helping Synergy Molding Technologies Prepare for Growth
From Automation Alley, tracked in The Axiovex Signal — headline, source, and date only; the item links to the publisher.
From the blog · Nov 11, 2026
We raised a Jetson Orin Nano's GPU speed limit in its firmware. Here are the measurements.
The GPU ceiling on our Jetson Orin Nano was set in firmware, not in the operating system. We edited 12 bytes of that firmware, raised the ceiling in 2 steps from 1020 MHz to 1173 MHz, and benchmarked everything before and after. Prefill got about 11% faster. One sustained workload got slightly slower, and we are publishing that too.
Featured
October 30, 2026
Michigan's September unemployment rate held at 5.0% — on a labor force that shrank by about 26,000 for the second straight month. The September-data edition, with trend, outlook, education and geography boards and the October WARN calendar.
Read article →The method
A constraint is a limit that is actually true: an opening that really exists, a course that really runs, a credential that takes as long as it takes, a lab with only so many stations, a person who can be in only one place at a time. Constraint-based planning means collecting those limits first — all of them — and building only the plans that satisfy every one at the same time. Not a plan built on averages. Not last year’s pattern with this year’s dates. A plan that fits the world as it is.
Feasible under the model — or honestly reported
Every plan this method produces is checked against all the documented limits at once. And when no plan fits, it says so — and names the constraints that bind, the limits actually in the way. Feasible here means feasible under the model, on the inputs as documented and dated; it is not a guarantee about the world. That turns a vague “it can’t be done” into a specific decision: change this limit, or change the goal.
Conflicts surface before commitments
A course that finishes after the job starts. A cohort larger than the room it trains in. Two employers counting on the same graduate. In ordinary planning, these turn up late, when they are expensive. Here they turn up on paper, while the plan can still change.
Trade-offs in the open
Real limits force choices: the same people, rooms and hours cannot serve two plans at once. The method makes those choices explicit — what was traded, for what, and who decided — instead of letting them hide inside a spreadsheet.
Re-solved when the facts change
Openings close. Schedules move. People drop out, and new people arrive. When a fact changes, the plans it touches are re-checked against the limits — so the plan in front of you still fits, instead of being one that used to.
What we don’t sell
We don’t sell seats — that is a college’s work. We don’t sell software off a shelf — that is a vendor’s. We don’t sell staffing — that is an agency’s. What Axiovex provides is the plan itself: the feasible plan hidden inside your real constraints, found and checked. Axiovex Arcturus is the program we are developing to run plans like these at program scale — it is in development, being proven through this work first. Organizations participate in Arcturus under program agreements; it is not a product, and it is not sold, licensed, or offered by subscription. And when there isn’t one, you get a straight answer instead of a hopeful one — here is the constraint that binds, and here is what changing it would take.
The projected constraint engine (PCE) — the software we are developing to do this checking at scale — is under development, and its guarantees are being validated against conventional optimization, simple matching and human coordination, using the same inputs and objectives. The discipline above is how we plan today; the engine is how it scales.
Axiovex is named for this: axio, a foundational truth, and vex — to shake. We shake the assumptions first, then plan on what survives. The full name story is in About.
One method, everywhere
Constraint-based planning is not one service among several. It is how Axiovex does each of its jobs — the limits differ; the discipline doesn’t. It is also the method behind our flagship project: a data pipeline of insights that benefit the Michigan economy as a whole, which we are building now. The recommendations it produces today are the ones in its published outputs and in our client engagements; the program it feeds will add its own as it is proven.
The pilot
The workforce pilot and employer planning
The pilot is Arcturus’s first cohort. One institution, 10–20 employers and 20–30 participants plan around a single occupation family. The plan has to fit the openings that exist, the courses that run, and the people who can actually take part — all at once, or it isn’t the plan.
The data
The Axiovex Signal and Michigan Workforce Intelligence
The pipeline’s public outputs are already live: Axiovex Signals, our daily board; The Axiovex Signal, our monthly newsletter; and Michigan Workforce Intelligence, our monthly report — published, dated and sourced, so a plan is built on this month’s limits, not last year’s. Axiovex is both: consultants and researchers doing this work today, and the builders and future managers of the program it adds up to: Axiovex Arcturus, a workforce and capability program for participating employers, universities, and workforce organizations, built on this method and fed by this pipeline. It is not sold, licensed, or offered by subscription.
AI work
AI readiness and adoption
The same discipline, applied to AI: start from a company’s real limits — its people, its data, its time, what it can afford to risk — and find the adoption plan that fits them, instead of a generic roadmap written for a company that doesn’t exist.
The build
How we build systems
When Axiovex builds, the requirements and the limits come first, and the build has to satisfy them. A requirement that cannot be met surfaces before anything ships — the same rule as the plans: feasible under the requirements as documented, or honestly reported as not feasible, with the constraints that bind named.
Our goal
Program scale, measured on real jobs.
Our goal is to put this method to work at program scale in Michigan — planning, and in time operating, workforce programs measured on whether people land in real jobs. The work connects across the wider ecosystem — employers, associations, workforce boards, educators, training operators — rather than depending on any single party, and it keeps proving itself in public through the monthly workforce report and Signals.
The pilot program
Program functions, developed through defined pilot scopes with participating institutions, employers, and workforce organizations.
01 · Institutions & employers
Education & employment
Education planning: map employer needs and future demand scenarios to skills, courses, credentials and cohort timing.
Student and graduate pathways: coordinate paid internships, co-ops, apprenticeships, skill bridges and graduate recruitment.
Continuing education: plan upskilling, internal redeployment and voluntary transitions as work changes.
02 · Companies & industry hosts
Company capacity introductions
In development — designed, not yet offered
Identify a possible fit between available machining, production, testing, laboratory or training capacity and another participating company’s stated need. No matching is happening today.
Arcturus can introduce companies that consent to be introduced. After the introduction, companies work together directly, in their own systems. Arcturus is not a broker and takes no part in any arrangement between companies.
03 · Regional hosts & ventures
Founder & startup attraction
In development — designed, not yet offered
Review bounded public patent, research, news and professional signals to identify ventures worth a closer look. No venture has been identified or relocated under this function yet.
Verify demand, readiness, IP access and interest in Michigan before building a conditional plan around customers, talent, facilities, suppliers and support.
We’re also exploring startup and growth packages: websites, corporate operating systems for day-to-day work, and practical help navigating Michigan’s business resources. These would support both new ventures and existing companies. The packages are exploratory: scope and eligibility are confirmed with each business before anything begins.
Start with one decision worth improving.
The proposed first pilot brings one institution, 10–20 Macomb-area employers and 20–30 participants around one occupation family. Scope, participation, data rights and delivery capacity must be confirmed before launch.
Discuss a first pilotEvidence in practice
Michigan workforce intelligence
Evidence for education and industry.
Our published workforce research and Signals board provide an initial evidence foundation. Figures retain their source periods; historical outcomes do not establish current job or resource availability.
READ THE WORKFORCE ANALYSIS →Constraint research & validation
Research with its limits in view.
Read the research record and its limits. The flagship’s engine and delivery model need their own technical comparisons, participant outcomes and paid renewal evidence.
READ THE RESEARCH RECORD →Training and private AI engineering can support a defined engagement. Our open AI 101 course materials remain available.
Independent projects by people we know and like. A link here is a friendly pointer, not a partnership, an endorsement, or a business relationship.
Market Intelligence: Rates, AI Supply Chains, Energy & Michigan
marketinteloutlook.com
Hire PCB, Firmware & Electronics Engineers
electrohire.com
Programs and institutions whose work connects with ours. A link is a pointer, not a partnership or an endorsement.
Macomb Community College
macomb.edu
Macomb Regional APEX Accelerator
macomb.edu
Michigan Alliance of APEX Accelerators
miapex.org
Michigan Works! Association
michiganworks.org
MEDC
michiganbusiness.org
Job Corps
jobcorps.gov
OU INC
oakland.edu/ouinc
Velocity
mivelocity.com
Michigan Manufacturing Association
mimfg.org
Aerospace Industry Association of Michigan
aiamnow.com
Networks Northwest
networksnorthwest.org
Axiovex Signals
Michigan’s numbers, and the field items that matter to the people who make things — from official sources and named publishers. The board updates when its sync succeeds; the snapshot stamp below is its freshness fact.
MANUFACTURING DIVE
Automakers using advanced tech to cut waste and costs, boost flexibility
November 13, 2026
NIST
Self-Employed? Solopreneur? NIST Has a Cybersecurity Guide Just for You
November 2, 2026
CISA
Siemens Building Automation and Energy Products
November 12, 2026
Snapshot updated November 15, 2026 at 8:27 AM ET. Headline, source, and date only — every card links to the publisher.
THE AXIOVEX SIGNAL — MONTHLY
What changed in Michigan industry and the field → one email, after the monthly economic releases, with the numbers, their vintages, and why they matter.
We sent a confirmation email to the address you gave us. Click the link in it — good for 48 hours — and everything you checked is confirmed. No click, no emails: that’s the deal.
Three separate lists — join only the ones you want, and leave any of them in one click, any time. We send a confirmation first — nothing arrives until you click it. No tracking pixels, no per-reader analytics. Privacy policy
About
Axiovex Systems is a Michigan company founded by Tristen Pierson and Rockson Arnold. We are developing constraint technology and coordinated services that connect education, employment and industrial growth.
We begin with a bounded problem, an accountable buyer and evidence that essential data can be obtained. Useful pilots can become repeatable services when results, delivery costs and customer renewal support expansion.
Axiovex is axio — from axiom, a foundational truth — and vex, from the Latin vexare: to shake, to agitate, to trouble. The name is the method. We shake the axioms — questioning the foundational assumptions until what is actually true falls out — then build on that ground. Every engagement starts there.
Tristen Pierson
Co-Founder & President
Rockson Arnold
Co-founder & Chief of Strategy
Questions
We are developing constraint-based planning and coordinated services for education and employment, company resource sharing, and founder attraction in Michigan. Engagements begin with qualification and a fixed scope; training and technical delivery are scoped separately where needed.
It’s planning that starts from the real limits — confirmed openings, the courses and credentials that lead to them, schedules, places, capacity and time — and builds only plans that satisfy all of them at once. When no plan fits, it names the limits that bind, so you know what would have to change.
The PCE is a proposed mechanism under development. Production projection guarantees and comparative advantage require validation against conventional solvers, simple matching and competent human coordination. Published research does not by itself establish readiness for every pilot.
The proposed education and employment pilot starts with one occupation family, one institution, 10–20 Macomb-area employers and 20–30 participants across student, graduate and incumbent-worker pathways. These are planning targets; launch requires confirmed commitments and capacity. Join the Arcturus pilot waitlist →
Access requires specific agreements. A documented vendor API does not grant access to a partner’s records. We start with named data owners, permitted uses and approved samples or exports; automated feeds follow when they improve a useful decision.
Agree a baseline and outcomes before launch. Track adopted education changes, paid starts and retained employment; qualified capacity work and total costs; or verified founder interest and subsequent Michigan activity. Targets are not findings, and an introduction is not a relocation.
Yes. AI 101 materials remain openly available, and a defined engagement can include training, private AI infrastructure or engineering. These capabilities support the agreed problem and do not substitute for verified employer demand or participant outcomes.
We are exploring packages that could combine a website, a corporate operating system for managing day-to-day work, and guidance on relevant Michigan business resources. The aim is to help new startups and existing companies make better use of available support. The packages are exploratory; scope, eligibility and access are confirmed for each business before anything begins, and no price or start date is published.
We are a Michigan company rooted in Macomb County. Our work here begins with a bounded local pilot before any wider expansion.
Axiovex Systems was founded in September 2026 by Tristen Pierson (Co-Founder & President) and Rockson Arnold (Co-Founder & Chief of Strategy).
Next step
Tell us the decision, who owns it, the data you can authorize and the outcome you need to improve.
Start a conversation
