A real problem at a working compounding pharmacy. Solved by local engineers — you.
5 minutes per compound. From 20 mg to 100 g in a single dose. Within ±3%. Thousands of times a day at MedRock alone.
MedRock Pharmacies identified hand-weighing as the bottleneck on their compounding floor. Every prescription that gets built on demand instead of pulled off a shelf starts at the same place: a pharmacist or technician at a scale, scoop in hand, working through one dose at a time. Tedious. Repeatable. Slow enough that the only way to scale that workflow today is to hire more people to stand at another scale. That's the problem this internship exists to solve. Not a school exercise. Not a thought experiment.
MedRock Pharmacies is a working compounding pharmacy. The compounds you'll watch get made are the ones MedRock fills for real patients every day. That's where the test runs happen. That's where the milestone validations happen. That's where Demo Day is.
MedDots is the engineering firm spearheading the mechanical development of the Lab Tech machine. Mechanical, electrical, controls, firmware — the full bench. They're not sponsors of this program; they're the engineers running it.
Itkan is the local hub making this possible — the workroom, the workshops, and the connections to schools and engineering programs across DFW and Chattanooga. Itkan isn't on the engineering panel; they're the reason a regional internship program can exist at all.
FIRST in Texas runs the robotics competitions and team programs that get high-schoolers their first real engineering reps — CAD, controllers, deadlines. They've been a backbone of regional engineering culture for years, and they're partnering with us to make sure this internship reaches the kids it's built for.
Anything your team designs, codes, prototypes, or documents during the cycle is assigned to MedDots and MedRock. That is the trade for the stipend, the pharmacy access, the engineering support, and the grand prize. You keep authorship credit, the experience, the relationships, and the right to talk about your role in interviews and applications. You don't keep commercialization rights to the machine. If that's not your trade, that's fair, and you shouldn't apply.
5 milestones spaced every two weeks. Each has a published acceptance checklist (what mechanically has to be working) and expected uploads (CAD, BOM, code, demo video).
Live, in-person, on the MedRock floor. The judge panel scores every team that's hit all five milestones.
Judges deliberate. Grand prize announced.
The spine of the cycle. Pass = stipend. Miss = catch up at the next visit. Hit all five and you're grand-prize eligible.
Build the part that physically moves powder — a mechanism that pulls a controlled amount out of the hopper (the bin the powder sits in). No scale or automation yet; the goal is a dose you can make bigger or smaller and roughly repeat by hand — and a doser that comes apart for cleaning. It's the foundation every later milestone builds on.
A focused milestone — fewer boxes than the others, but the one that matters is a big leap: weighing tolerance jumps from ±20% (rough, hand-actuated in M1) to ±3%. Add the controller and scale so the machine weighs powder as it dispenses and stops itself — "closed-loop" control, where the scale's reading feeds back to the controller, replacing hand-counting and guessing on time. This is where rough dispensing becomes real precision.
Push the single-doser machine hard: prove it's accurate across the whole weight range, handles every powder on the test list — not just the easy free-flowing ones — and repeats run after run. This is the rigor checkpoint before you add more dosers, and where the machine starts driving itself from a basic on-device screen.
Now more than one powder per recipe. Several dosers feed one scale and build a compound (a mixture of ingredients) where every amount adds up correctly — the machine tracks which powder is loaded in which doser, and the on-device interface grows to build recipes and manage settings.
Full autonomy: an operator loads the powders, enters a recipe on the device, hits go, and walks away while the machine makes the compound start to finish — logging every step. This is also the Final Demo Day — the judge panel scores your machine live.
We're picking up to 5 teams out of a maximum of 20 applications. We want teams that:
The pitch. Three short prompts on why your team— not your engineering approach, not your design ideas. We're picking the people, not the proposal.
5 teams. 12 weeks. One real machine going onto a working pharmacy floor. Form a team, write a pitch, send it in.
The bar: discipline coverage across 4–6 people, real prior engineering work, the resources you already have access to, and a pitch that makes the case for your team specifically.
