Infinite Technology System

The number on the screen was 98.7%.

Dhiraj stared at it for four seconds without moving.

Then he checked the sensor calibration. Then the diagnostic software version. Then the physical connections at the base of the monitoring rig, one by one, because his hands needed something to do while his brain finished arriving at the conclusion his eyes had already reached.

The number didn’t change.

98.7%.

He sat back in the workshop chair and looked at the prototype sitting under the fabrication lights — compact, gray-cased, no exposed wiring, no visible improvisation. It looked like something that had been designed by a team with a budget and a timeline and actual equipment. It did not look like something assembled between midnight and 3 a.m. on hostel floor tiles surrounded by dismantled power regulators and a mosquito that had made the tube light its personal project.

The diagnostic software updated once, quietly, with the indifference of a system that didn’t understand what it was measuring.

Thermal regulation efficiency: 98.7%.

Discharge stability: nominal.

Structural degradation: negligible.

He exhaled through his nose.

Modern commercial systems were nowhere near this number. Research labs with eight-figure budgets published results in the low eighties and held conferences about it. The theoretical ceiling that three separate papers had argued was physically insurmountable sat at roughly 91%, under ideal conditions, using materials that cost more per gram than this entire workshop’s equipment budget.

Version four was running at 98.7%.

In a deteriorating university workshop with cracked floor tiles, a broken exhaust fan, and a ceiling light that had been flickering since February.

Dhiraj looked at the prototype for a long moment.

Then, because there was no one in the workshop to see it, he allowed himself exactly three seconds of something that wasn’t quite happiness but was the structural precondition for it — the specific, quiet recognition that the thing you’d been building toward was real. Not theoretically. Not on paper. Real, on a workstation, holding a number that had no business existing.

Three seconds.

Then he started documenting.

---

> TECHNOLOGICAL MILESTONE DETECTED.

The System’s voice arrived without preamble, the way it always did — no warmth, no announcement, just information delivered with complete indifference to how it landed.

> PROTOTYPE CLASSIFICATION UPDATED.

> PREVIOUS: ADVANCED RESEARCH SYSTEM.

> CURRENT: PRE-COMMERCIAL STRATEGIC INFRASTRUCTURE TECHNOLOGY.

> TECHNOLOGY POINTS AWARDED: +150

> SYNCHRONIZATION: INCREASING.

> NEW PATHWAYS: PENDING ASSESSMENT.

Dhiraj read through it twice.

*Strategic infrastructure technology.*

Not a research breakthrough. Not an advanced prototype. The System had drawn a line between those categories and placed version four on the other side of it, and the distinction mattered in ways that took him a moment to fully map.

Research breakthroughs could be ignored. Dismissed. Left in papers that took years to replicate and longer to matter.

Strategic infrastructure technology was different. That was the category of thing that determined who controlled energy grids. Who won manufacturing contracts. Who governments called when supply chains failed. The category of thing that attracted people who didn’t introduce themselves by their real names.

He looked at the prototype.

Still running. Still quiet. Still holding 98.7% with the patient indifference of something that had never doubted itself.

Outside the workshop window, Mumbai’s evening skyline was doing its usual negotiation with the power grid — one section flickering near the eastern district, load balancing kicking in two seconds later, everything recovering before most people noticed it had been anything other than fine.

Version four could fix that.

Not just fix it. Change the underlying architecture that made it a problem in the first place. Grid storage, AI compute budgets, EV charging infrastructure, remote industrial operations — the downstream dependencies kept extending outward the longer he thought about it, each one unlocking the next, a cascade of second-order consequences that started with a number on a diagnostic screen and ended somewhere he couldn’t fully see yet.

His fingers had stopped moving over the documentation.

He made himself continue.

---

Rohit arrived at half past six with two cold coffee bottles sweating in the heat and the particular expression of someone who had walked a long distance and was reconsidering the decision.

He stopped in the workshop doorway.

Looked at the prototype.

Looked at Dhiraj.

Looked back at the prototype.

"That’s not version three."

"No."

"It looks different."

"Yes."

Rohit walked in slowly, moving around the workstation the way you moved around something you weren’t sure was safe to approach directly. He had the expression engineers made when a thing didn’t fit neatly into what they already understood — not disbelief exactly, but the moment before disbelief, when the brain was still trying to locate the error.

"The last one looked like something you built," he said.

"This one looks like something that was built."

Rohit stared at it for another second. "Yeah."

He set one of the coffee bottles beside Dhiraj without being asked. The condensation left a ring on the workstation surface immediately. Neither of them mentioned it.

"How bad are the numbers?" Rohit asked.

"The opposite of bad."

A pause.

"How opposite."

Dhiraj turned the monitoring display so Rohit could see it.

Rohit looked at the screen. Then at Dhiraj. Then back at the screen, with the particular stillness of someone whose mental model of the situation had just been quietly demolished and was taking a moment to rebuild.

"98.7," he said.

"Yes."

"That’s not — Dhiraj, that number doesn’t —"

"I know."

"The theoretical ceiling according to the Singh paper is —"

"91.4. Under ideal conditions."

Rohit sat down on a nearby stool without appearing to decide to. He looked like a man who needed to be horizontal but was settling for seated.

"You’re running 98.7," he said slowly, "in a room where the exhaust fan is broken and the floor tiles are cracked."

"The tiles aren’t load-bearing."

"That’s not the point I’m making."

Outside, a bike horn sustained itself for several seconds longer than any legitimate traffic purpose required. The ceiling fan overhead continued its uneven rotation, one blade slightly heavier than the others, clicking softly on every pass.

Rohit picked up his coffee. Set it down again without drinking it.

"You actually did it."

Not a question. Not congratulations exactly. Something more careful — the tone of someone who had believed in the thing and was now confronting what believing in it actually meant.

Dhiraj turned back to the documentation.

"Version three proved the concept. Version four proves it can be built." He paused. "Those are different problems."

"I know they’re different problems." Rohit looked at him. "I’m saying you solved both of them."

The workshop was quiet for a moment.

Somewhere down the corridor, someone dragged a chair across the floor with a sound like a small structural protest. The monitoring software updated again — numbers unchanged, consistent, holding — and then returned to its patient silence.

"Kulkarni asked about you again," Rohit said.

The shift in his voice was small but there.

Dhiraj looked over. "When."

"This afternoon. Faculty corridor." Rohit turned the coffee bottle in his hands. "He asked whether you were still doing independent energy research. Used that exact phrase."

"Not just asking then."

"No." Rohit met his eyes. "He was checking whether the answer had changed."

---

The anonymous email was still sitting in Dhiraj’s phone when he got back to the hostel.

He’d read it four times since it arrived the night before. He read it a fifth time now, sitting on the floor with his back against the bed frame, in the same position he’d been in when he was assembling the conductive layers at 3 a.m. — a posture that had apparently become load-bearing.

*Revolutionary technology attracts attention faster than expected. Be cautious about corporate dependency.*

No greeting. No signature. Routing metadata professionally obscured, which was its own kind of signature.

He checked the timestamp again. It had arrived eleven minutes after he’d sent the three-word reply to Vikram Malhotra’s meeting request.

Eleven minutes.

Someone was watching his communications fast enough that the email had functioned as a response. Not to what he’d sent — to the fact that he’d sent it at all.

> EXTERNAL OBSERVATION: CONFIRMED.

"Voltaris?" Dhiraj asked.

> INSUFFICIENT DATA.

"Consortium?"

> INSUFFICIENT DATA.

"You’re going to say that every time the answer makes you useful."

The System did not acknowledge this.

He set the phone down on the floor beside him and looked at the ceiling. The hostel fan clicked overhead in its uneven rhythm, patient and indifferent. Through the wall, someone was watching something on their phone at a volume that suggested they’d stopped caring whether the wall was there.

Three possibilities. He’d been through them already.

Voltaris had internal divisions — someone inside the company who thought the partnership was moving too fast, or in the wrong direction, or toward the wrong person.

An independent observer had been tracking the energy anomaly from the university building and had connected enough dots to find a name.

Or something larger was already in motion and this was its first visible edge.

The third option was the one he kept returning to.

Not because it was the most likely. Because it was the one that made the other two make sense as well.

> HOST DECISION PATTERN: NOTED.

"What does that mean."

> YOU ARE THINKING STRUCTURALLY RATHER THAN REACTIVELY.

"Is that good."

> IT IS APPROPRIATE.

He looked at the prototype on the desk. The blue indicator lights pulsed slowly through the casing seam — soft, even, the same rhythm they’d held since the first successful discharge test. The adhesive residue near the lower vent caught the light in a way he’d stopped noticing.

Version four had been classified as strategic infrastructure technology approximately three hours ago.

Which meant the question was no longer whether someone would come for it.

The question was who arrived first, and what they thought they were arriving for.

The meeting with Vikram Malhotra was tomorrow.

He’d confirmed it before he understood what he was confirming. Before the anonymous email. Before Kulkarni’s second inquiry. Before the System had reclassified the prototype and quietly moved everything into a different category.

He thought about canceling it.

Ran the logic.

If Voltaris was the source of the email — warning him away from corporate dependency while simultaneously scheduling a meeting — then canceling told them he’d made the connection. Which was information.

If Voltaris wasn’t the source, then someone else already knew about the meeting. Canceling didn’t make him invisible. It just made him reactive.

He left the confirmation in place.

Then opened the notebook to a fresh page and wrote across the top in the careful handwriting he used when the thing he was writing mattered:

*What does Voltaris need badly enough to send an anonymous warning about itself?*

He stared at it for a moment.

Crossed out *anonymous warning about itself.*

Wrote: *What does Voltaris need badly enough that someone inside it is afraid of what happens next?*

Better.

He put the pen down and looked at the prototype.

98.7%.

It hummed in the corner of the room, soft and patient, throwing its slow blue pulse across the wall in the dark. The most advanced energy storage architecture on Earth, sitting on a hostel desk between an empty instant noodle cup and a charging cable that had surrendered its original shape sometime in October.

Everything was about to get harder.

He’d known that since the discharge test. Since the System had started reclassifying things. Since Rohit had sat on a workshop stool with the expression of someone recalibrating what believing in something actually cost.

He just hadn’t known how fast.

Outside, the city held its margins through the evening — load balancing, backup systems, the accumulated expertise of millions of people managing small failures before they became large ones. The particular ordinary of a civilization running at its own limits and calling it normal.

One section of skyline flickered.

Recovered.

Held.

Dhiraj closed the notebook.

He didn’t sleep for another two hours. Not because he was afraid — he’d catalogued that and set it aside. Because somewhere underneath the dread and the logistics and the anonymous email and Kulkarni’s careful questions, the number was still there.

98.7%.

Real. His. Impossible.

And no matter what came next, no one could make it not have happened.

*That* was the thing worth holding onto.

Everything else was just what it cost.

---