The Workshop #
I was, at this point in my life, very good at alchemy. I had been very good at alchemy for approximately twelve hundred years. Herbs did what I told them to do. Reagents combined in predictable ways. The formulas were old friends — I knew their moods, their preferences, the particular way a firebloom needed to be crushed rather than sliced, the exact moment when a transmutation would tip from almost to done.
Alchemy had taught me precision. It had taught me that method outlasts mood. It had taught me, during those ten days of arcanite transmutation for a dracthyr warrior I’d met on forge steps, that sometimes the best thing you can do for someone is simply show up and follow the formula.
But alchemy was not, as Brasskeese had pointed out with characteristic bluntness, enough.
“You’ve got the ingredients part down,” she said one evening, watching me decant a fresh batch of healing potions in the workshop I’d carved out of a corner of the common room. “You’ve had the ingredients part down for centuries. But ingredients are passive. They sit there until you do something with them. Engineering is different. Engineering is active. Engineering doesn’t wait for you to be ready. It just works.”
She placed something on the workbench between us. It was small. Brass. Covered in goblin warning labels. It looked like a cross between a pocket watch and a tiny bomb — which, I would later learn, described approximately seventy percent of goblin inventions.
“What is it?”
“Automatic shield generator. Wrist-mounted. You strap it on, you press the button, it throws a barrier around whoever you’re pointing at. Doesn’t require mana. Doesn’t require concentration. Doesn’t require you to be conscious, actually, which I tested. Unintentionally.”
I picked it up. It was heavier than it looked. “You built this?”
“I built the prototype. The prototype exploded. This is version four. Version three also exploded but less dramatically. Engineering is a process.”
I turned the device over in my hands. A shield generator. No Light required. No Shadow required. Just — mechanics. Gears and springs and something that hummed faintly when I held it to my ear.
“Teach me,” I said.
Brass grinned. It was the grin of someone who had been waiting to be asked.
The Philosophy #
The first lesson was not about gears. It was about failure.
“Goblin engineering,” Brass said, spreading a set of blueprints across the table — diagrams that looked less like technical drawings and more like the excited scribbles of someone who had been struck by inspiration mid-explosion, “is not about getting it right the first time. It’s about getting it wrong in increasingly useful ways. You blow something up, you learn why it blew up, you build it again with one less thing that explodes. Repeat until the explosions are manageable.”
“That sounds… inefficient.”
“It sounds inefficient to someone who’s spent twelve centuries perfecting alchemy formulas. But here’s the thing — alchemy is a closed system. You put in X, you get out Y. Engineering is an open system. You build something, you see how it fails, you build it better. Every failure teaches you something alchemy’s perfect formulas never will.”
She tapped the shield generator. “Version one stunned the user instead of protecting them. Version two worked but only for three seconds. Version three worked for ten seconds and then caught fire. Version four works for thirty seconds and only smokes a little. That’s progress. That’s engineering.”
I thought about the Discipline path — the constant adjustment, the thousand tiny corrections, the balance that was never still. “That’s… actually a lot like what I do. Just with more explosions.”
“Everything’s better with more explosions,” Brass said. “Now. Let’s start with something simple. This is a spanner. You use it to hit things that aren’t working. Also to turn bolts. But mostly hitting.”
The First Build #
My first project was a healing dispenser.
Not a potion — I had been making potions for centuries. A dispenser. A mechanical device that would hold a healing potion, monitor the wearer’s vitals, and administer the dose automatically when needed. No priest required. No healer on standby. Just — a machine, doing what I would do, while I was somewhere else.
The first version sprayed healing potion directly into my face.
“This is good,” Brass said, handing me a towel.
“It sprayed potion into my face.”
“Which means the spray mechanism works. You’ve just got to point it the other way. That’s a directional problem. Directional problems are easy.”
The second version pointed the right way but administered the potion continuously, draining the entire reservoir in four seconds. Vyrneese, who had volunteered to test it, stood very still with healing potion dripping from her chin and said, “I do not think I am this injured.”
“Overhealing,” Brass noted. “Tone down the flow rate.”
The third version worked. When Vyrneese — still damp, but game — held it to her arm, the device detected her vitals, waited, and when nothing was wrong, did absolutely nothing. Brass clapped me on the shoulder hard enough to stagger me.
“Now you’re thinking like an engineer. The goal isn’t to always be doing something. The goal is to do the right thing at the right time. Sometimes the right thing is nothing. That’s the hardest part to learn.”
I looked at the device — small, brass, covered in my own warning labels now — and felt something shift. Alchemy had taught me formulas. Engineering was teaching me judgment. When to act. When to wait. When to let the machine do the work.
It was, I realized, not so different from the Discipline path.
The Long Night #
Late one evening, long after Brass had gone to bed and Vyrneese had retreated to her room with Thunderfury propped against the wall like a nightlight, I sat alone in the workshop. Around me: gears, springs, three versions of the healing dispenser, a stack of goblin engineering manuals with titles like So You’ve Decided to Explode, and a half-finished project that Brass had called “ambitious” — which, from Brass, was either a compliment or a warning.
I was building a portable barrier generator. Not as elegant as Power Word: Shield — nothing mechanical ever would be. But Power Word: Shield required me. Required my focus, my mana, my presence. This device, if it worked, would shield anyone who wore it, whether I was there or not.
The tea was cold. Of course it was cold. Some things don’t change.
But something had changed. A year ago, I had been a priestess who happened to know alchemy. Now I was a priestess who knew alchemy and was learning engineering and lived in a house with rules and a landlord who hit things with spanners and a dracthyr who tested my inventions without complaint. A year ago, I had been alone. Now I had — not a warband, not yet, not the word for it — but something. Something that felt like the beginning of an answer to a question I hadn’t known I was asking.
I picked up the spanner. I looked at the barrier generator. I thought about failure — the useful kind, the kind that teaches you something — and I got back to work.
The Point #
Vyrneese found me the next morning, still in the workshop, oil on my robes and a completed barrier generator in my hands.
“Did you sleep?” she asked.
“No.”
“Is the thing done?”
I held it up. Small. Brass. Warning labels. A faint hum when you held it to your ear — the same hum as Brass’s shield generator, the one that had started all of this.
“It’s done. It might explode. Brass says that’s normal for version one.”
Vyrneese took it from me, gently, the way she handled everything — with the particular care of someone who knows exactly how strong she is and exactly how fragile everything else can be. She turned it over in her hands. The morning light caught the brass casing and threw small golden reflections across the workshop walls.
“A year ago,” she said, “you were a priestess on forge steps who helped me because you could. Now you build machines that protect people while you sleep.”
“That’s the idea.”
“Is this what Brass meant? About systems?”
I looked around the workshop — the failed prototypes, the successful ones, the manuals with their enthusiastic titles, the spanner still warm from use. “I think so. Alchemy taught me precision. Engineering is teaching me something else. That you don’t have to be perfect. You just have to keep building. Keep failing. Keep starting over. The systems get better. You get better. And someday —” I gestured at the barrier generator in her hands — “someday they work. Even when you don’t.”
Vyrneese was quiet for a moment. Then she said, very seriously: “I am glad you did not sleep. This is a good machine.”
“Thank you, Vyr.”
“I will test it. After breakfast. In case it explodes.”
“That’s… very prudent.”
“I am learning prudence. Brass says it is one of the rules.”
I laughed. It was the first time I had laughed in the workshop — really laughed, not the quiet smile of someone working alone, but the full, unexpected laugh of someone who had found something they didn’t know they were looking for.
Outside, the dwarven quarter was waking up. Somewhere in the kitchen, Brass was probably counting copper and pretending not to care. Somewhere in the house, the tea was getting cold. And somewhere in the shape of things to come, a demon hunter was waiting with a grudge and a blindfold and a question I wouldn’t be ready to answer.
But not today. Today, there was a barrier generator. And breakfast. And a dracthyr who had learned prudence.
It was enough.