For 30 advance early chapters : p atreon.com/AutumnXd
The senior team stayed in the lab analyzing data. Ryan copied the experimental results to a hard drive and took it to Patricia.
She looked up from her desk, surprised. "I thought you'd be in the lab all day."
"Experiment's done."
She checked the time. Not even noon. "Already?"
"Three successful dual connections. All data on the drive. Send it up."
Patricia took the drive without further questions. She'd learned not to ask for details she wasn't cleared to receive. The project's classification structure was layered: she knew enough to manage logistics but not enough to understand the science. That was deliberate, and she respected it.
Within the hour, a helicopter lifted off from the facility's landing pad. Ryan heard the rotors from his room and made a mental note to ask Patricia whether that helicopter was funded from his project budget.
He sat down to study the plasma cannon specifications and was immediately interrupted by a knock.
Kyle stood in the doorway looking like a man who hadn't slept in a week. His hair was matted. His eyes were sunken. His glasses sat crooked on his nose. He looked like he'd crawled out of a research paper and couldn't find his way back.
"You okay?"
Kyle's voice was hoarse. "I have something to show you. A modification proposal for the neural link's power consumption profile. For the firefighting mech."
He held out a hard drive.
Ryan took it. Opened the files. Started reading.
Within the first few pages, his eyebrows went up.
Kyle had done real work. Not perfect work. The proposal had errors, places where he'd oversimplified a subsystem interaction or miscalculated a load parameter. But the overall approach was sound. He'd identified a genuine inefficiency in the neural link's power draw and proposed a restructuring that could reduce energy consumption by a meaningful margin.
For a master's student who'd been studying the technology for three months, it was remarkable.
Ryan pulled Kyle to the desk and started walking through the corrections.
"This backup system you deleted here. You can't remove it. Without the redundant pathway, the pilot loses the ability to restart the neural connection after a system fault. In a firefighting scenario, that could be fatal."
Kyle nodded, taking notes.
"But your core idea about rerouting the primary signal path through a lower-resistance channel is correct. You just need to preserve the backup while optimizing the main line."
Ryan continued through the document, correcting, refining, confirming. Kyle absorbed everything with the desperate gratitude of someone who'd poured months of work into a single document and was finally hearing that it wasn't garbage.
The helicopter reached Aegis headquarters. The data was decrypted, distributed, and reviewed. A meeting was called within two hours.
"Another experiment result from the asset," the briefing officer announced, projecting the summary on the wall screen.
Result: Three dual-person neural connections. All successful.
The room stared.
"Three separate pairs?"
"Different subjects each time?"
"Maximum sustained connection: over thirteen minutes."
Someone pulled up the monitoring video. The room watched the experiment from start to finish. The sensor caps activating. The reaction solutions cycling. The brothers sitting motionless, eyes closed, neural patterns merging on the display.
The memory-sharing data drew the most attention. The documentation described how connected subjects could perceive fragments of each other's memories during the drift. Not full access. Not detailed recall. But direct observation of another person's neural imprints, surfacing involuntarily during the connection.
The implications were obvious. And enormous.
"He says he's not pursuing the memory-reading applications further," the briefing officer noted. "He's leaving that for us to develop."
"Generous of him."
"Pragmatic. He wants to focus on the mech. We get the spin-off research."
"Speaking of which." Someone leaned forward. "If the dual connection works this cleanly, the three-person drift he needs for the Jaeger becomes viable. And if the drift works, the plasma cannon project becomes relevant. Because a Jaeger with a working drift system and a working plasma weapon is no longer a theoretical concept. It's a platform."
A pause while the room recalibrated.
"So we're approving the plasma cannon project?"
"I think we have to. The neural link validates the drift. The drift validates the Jaeger. The Jaeger validates the cannon. It's a chain, and every link is holding."
"What about the budget? The initial hundred million won't cover weapons research."
"He hasn't asked for more yet."
"He will."
A few people laughed. They were getting used to Ryan Mercer's rhythm: deliver a result, immediately request the next investment, deliver again before the ink dried on the approval.
"Want to bet whether he produces results with the first tranche of funding again?" someone asked.
Nobody took the bet.