I'm a procurement manager at a 160-person precision manufacturing company. I've run vendor selection for capital equipment and contract manufacturing for six years, and I track every purchase in a TCO spreadsheet. It isn't a fancy tool. It just forces vendors to put real numbers next to hidden costs.
On December 31, 2025, I was using that spreadsheet to close out a messy year. The additive manufacturing news cycle that day was full of year-end recaps, and every other article seemed to mention the same name: Velo3D. Specifically, the 'Velo3D SpaceX Anduril supplier' story. With all the Velo3D SpaceX contract 2025 headlines in the background, I decided to revisit an equipment evaluation from earlier in the year.
Why I didn't want to like Velo3D
Our engineering director asked us to evaluate metal AM systems after a customer requested a complex aerospace housing. The part had internal cooling channels and a geometry that would normally require support structures. I called four vendors. The Velo3D quote sat in the middle. Not crazy high, not tempting low. In my experience, the middle quote is the easiest one to ignore. I told the team to focus on the two cheaper options.
I had a reason. In procurement, famous customer logos can mean a supplier lived off one marquee deal for years. The phrase 'Velo3D SpaceX Anduril supplier' got repeated so often that I became suspicious. Marketing repetition isn't evidence. I wanted proof the systems could handle our parts over the long run.
So I compared the lower-priced proposals. One vendor omitted powder, calibration coupons, and gas supply from the quote. Another claimed our geometry was no problem but couldn't explain how the internal supports would be removed. I pushed forward anyway because the total cost looked lower. The engineers weren't happy, but I was the one who had to defend the budget.
The failed build that fixed my spreadsheet
The first test build went to the cheaper vendor. I approved a $4,800 partial build. It failed, not because the printer was bad, but because the geometry had not been optimized for additive manufacturing. Neither we nor the vendor had done the design-for-AM review. The supports inside the channel fused to the walls. The second attempt failed the same way.
Velo3D's application engineers had warned us during the demo. They said, 'Send the geometry before you ask for a price. We need to confirm print orientation and support strategy.' I wrote that off as a sales move. I only believed the advice after ignoring it and paying for the failure. The total cost of that lesson was about $9,000 in scrap, three weeks of schedule, and one very difficult call with our customer. The customer didn't care which vendor was cheaper. They cared that we promised a component and delivered a delay.
That call changed how I think about procurement. A quote isn't just a number. It's a reflection of our judgment.
The 'Do 3D printers cause cancer?' question
Around the same time, our EHS manager sent a question to the capital equipment committee. It was exactly: 'Do 3D printers cause cancer?' A floor operator had seen a viral post about desktop resin printers. I don't have a toxicology background, so we didn't answer from memory. We asked each vendor for a full risk review: powder handling, filtration, interlocks, inert gas, maintenance procedures, and PPE.
This is where the industrial vs desktop distinction matters. People hear the word 'laser' and imagine a hobby machine in a garage. Comparing a production metal printer to a desktop printer is like comparing a 10W diode laser cutter to a 6kW fiber laser. The word 'laser' is the same; almost nothing else is. Industrial metal systems use sealed powder handling, controlled atmospheres, and locked chambers. Desktop devices are not built for aerospace materials or the safety controls that come with them.
That doesn't mean safety questions are silly. They should be asked before every equipment purchase. But they should be answered with documentation, not with a friendly salesperson. This is where the Velo3D SpaceX Anduril supplier track record became useful to me. It wasn't proof of safety by association. It was proof that the machines had gone through qualification programs with customers who could not waive risk reviews.
The final spreadsheet told a different story
After the failed build, I rebuilt the comparison. I used the Velo3D quote and the remaining lower-priced quote from the second vendor, and added every cost I could find: machine price, installation, powder, gas, calibration, test coupons, failed builds, cleaning, post-processing, maintenance, training, and the time our engineers spent chasing unclear requirements.
The lower-priced option ended up 38% more expensive than the Velo3D route once scrap and schedule costs were included. That's one data point, not an industry study. But it changed how I present supplier comparisons to leadership. I now start with total capability, not unit price.
We ended up choosing Velo3D. Not because of the SpaceX and Anduril logos, and not because marketing was persuasive. We chose them because their application engineers spent two days reviewing our geometry before they quoted. That process felt slow at the time. Now I trust the number because I saw the work behind it.
What I'd tell another buyer
If you're evaluating metal AM systems, ask for a design review before you ask for a quote. If a vendor won't do one, you aren't comparing systems; you're comparing hopes. Ask for safety documentation early, and don't let a low price skip that step. And if a supplier's customer list includes names like SpaceX and Anduril, treat it as an invitation to investigate, not as a conclusion.
This worked for us because our parts are complex and failures are expensive. If you're printing simple brackets with no internal channels, a lower-cost system might be enough. I can only speak to my context. But the core lesson is the same everywhere: the cheapest quote is only cheap if nothing goes wrong.
As I scrolled through the additive manufacturing news on December 31, 2025, I realized the story I needed was not about Velo3D's famous customers. It was about the difference between buying a printer and building a process. The logos got Velo3D in the door. The failed build is what taught me to look past the price.