About MCG Labs: AI Consulting and Engineering
MCG Labs is a subsidiary of MCG LLC (Mercury Consulting Group), offering AI engineering, private AI hosting, and AI consulting. This page is about the consulting side: AI consulting that ends in working systems, not slide decks, and about the engineer who leads the practice.
How We Work
The same four rules apply whether we are building, hosting, or advising. They are what keeps an AI consulting engagement honest.
Source-grounded
Recommendations are checked against the datasheet, the codebase, the measurement, or the vendor's own documentation, not against a model's confidence or a generic best-practice list.
Verified end to end
A consulting deliverable is treated as atomic. It ships when every claim in it has been checked, or it does not ship yet.
Engineer in the loop
The person advising you is the same person who would build it. No hand-off between a strategy team and an implementation team that never talk.
Private by default
Where data or model access matters, we default to private hosting on hardware you control, not a shared cloud endpoint, unless you tell us otherwise.
AI Consulting, Done by Engineers Who Build
Most AI consulting engagements end at a strategy document: a ranked list of use cases, a maturity model, a vendor comparison. None of that tells you whether the thing actually works on your data, with your constraints, run by your team. MCG Labs' AI consulting practice is built around closing that gap. An engagement starts with a plain conversation about what you are trying to do and what is actually blocking you, moves into a scoped working session with the engineer who would build the system, and ends with something runnable: a pilot, a prototype, a decision document backed by a real test, not a guess.
That is the honest framing we give every prospective client up front. AI engineering consulting from MCG Labs is not strategy theater. If your problem needs months of sustained build work, the consulting engagement scopes it and then hands off cleanly into our augmented engineering pillar. If the constraint is that your data cannot leave your building, the conversation naturally moves into private AI hosting. If what you actually need is a shippable product with AI built into it, that is full-stack product work. AI consulting, in our shop, is the front door, not a separate product with its own incentive to keep billing.
We will also tell you when AI is not the right tool for your problem. An AI consulting firm that only ever recommends more AI work is not giving you advice, it is giving you a sales pitch.
Consulting ends when there is a working system, or an honest answer about why not. There is no third outcome we are willing to bill for.
Led by Engineers, Not Project Managers
Bil Herd, Principal Engineer
Bil Herd is a veteran hardware engineer with four decades in production electronics, promoted to Senior Hardware Design Engineer at Commodore Business Machines by age 24, where he designed the Commodore 128. He has been a regular Hackaday contributor since 2015 and now leads engineering at MCG LLC and MCG Labs, applying the same build-it-and-verify-it discipline to AI consulting and AI engineering that he has applied to hardware for forty years.
Read Bil Herd's full biography on mcgllc.com →
Sophi Kravitz, Testing, Analysis and Data Correlation
Sophi Kravitz is an expert electrical engineer and creative technologist who leads testing, analysis, and data correlation at MCG Labs. She holds a BSEE from the State University of New York and spent nearly seven years at Hackaday, most recently as Director of Product, where she ran flagship programs including the Hackaday Prize. Earlier in her career she worked as an engineer in battery and thin-film solar research, where measurement, test, and the analysis of large result sets were the daily work. At MCG Labs she owns the test-and-verify side of augmented engineering: designing the tests, running the analysis, and correlating large sets of measured data back to the design, so that what the AI produces is checked against what the hardware actually does. She is also an installation artist whose work pairs sculpture with technology.
Built on a Hardware Engineering Pedigree
MCG Labs did not start as a software shop that bolted on an AI story. It is the AI engineering arm of MCG LLC (Mercury Consulting Group) ↗, a hardware engineering practice that does architecture, PCB design and layout, and engineering-grade 3D printing for production electronics. That background is why the AI consulting work here looks the way it does: scoped, verified, engineer-owned, and allergic to claims that have not been checked against a real source. The same discipline that keeps a production board from coming back with a bad footprint rotation is the discipline that keeps an AI consulting recommendation from being a confident guess.
If your engagement turns out to need hands-on build time rather than advice alone, it continues with the people who did the consulting, across any of the three pillars: Augmented Engineering, Private AI Hosting, or Full-Stack Product. You can also see the discipline applied end to end in our case studies.
Frequently Asked Questions
What is AI consulting?
AI consulting is advisory and hands-on engineering work that helps an organization decide where AI genuinely helps, scope a pilot, and build the first working version, rather than a slide deck of recommendations. At MCG Labs it is delivered by the same engineers who do the implementation work, so the advice is grounded in what can actually be built and verified.
What does AI consulting cost?
It depends on scope. A short assessment engagement costs less than a multi-week build-and-verify engagement. We scope each engagement after an initial conversation about what you are trying to do, and quote a fixed scope rather than open-ended hours. There is no published rate card because the right engagement shape varies by problem.
How is MCG Labs related to MCG LLC?
MCG Labs is a subsidiary of MCG LLC (Mercury Consulting Group). MCG LLC is the hardware engineering practice: architecture, PCB design, and engineering-grade 3D printing. MCG Labs applies that same engineer-in-the-loop discipline to AI, offering AI engineering, private AI hosting, and AI consulting.
Talk to an engineer about AI consulting.
Tell us the problem. We will tell you honestly whether AI consulting, or one of our build pillars, is the right next step.
Email info@mcgllc.com