The contract was seventeen pages long, which seemed excessive for a startup with four people and a rented office in Zhongguancun. But the venture capital firm that had drafted it—apparently anticipating future funding rounds and investor requirements—wanted everything specified: equity vesting schedules, intellectual property assignment, non-compete clauses, performance milestones.
Chen Wei read it three times, understanding perhaps 60% of the legal terminology on the first pass, 85% on the second pass, and asking the system to translate the remaining 15% into comprehensible language on the third.
SYSTEM TRANSLATION (Article 6, Section 3):
"In the event of a funding round where external investors acquire an equity stake, your 2% ownership will be diluted proportionally. Your vesting period is four years with a one-year cliff (meaning if you leave before 12 months, you forfeit all equity). After cliff, equity vests monthly. This is a standard venture capital structure."
"So if the company gets acquired before my vesting cliff, I get nothing?" Chen Wei asked.
SYSTEM CONFIRMATION: Correct. This is a risk inherent to startup equity. An alternative would be full immediate vesting, but investors typically require a cliff to ensure founder commitment. This is a negotiated norm.
He signed the contract on January 23rd, two days before the official PhD enrollment process began.
The PhD part was almost anticlimactic by comparison. Professor Zhang had prepared all the necessary paperwork, coordinating with the graduate admissions committee to waive the typical GRE requirement (Chen Wei's publication and research record were deemed sufficient evidence of capability). The enrollment fee was paid. The advisor relationship was formally established.
By late January 2026, Chen Wei had officially transitioned into a new identity: simultaneously a startup CTO and a PhD candidate, with the expectation of managing both roles with equal dedication.
The first real test of that dual identity came at the Qiming Battery Materials office on February 2nd.
Dr Chen from CATL arrived with a manufacturing reality that none of them had entirely anticipated.
The meeting room was minimalist—three people (Li Na, Dr Chen, Chen Wei) around a borrowed conference table with a laptop displaying scaled-up synthesis protocols. Dr Chen had worked on battery manufacturing long enough to understand the gap between "works in the lab" and "works in production."
"Your synthesis protocol," Dr Chen said, pulling up his analysis, "scales linearly up to about 500 gram batches. Beyond that, your yield drops approximately 8-12% per 500 gram increment."
"Why?" Chen Wei asked immediately.
"Heat distribution," Dr Chen explained, displaying a thermal model. "In your lab synthesis, the furnace maintains uniform temperature throughout the sample. In a 2-kilogram batch—which you'd need for pilot production—the thermal gradient becomes significant. The centre of the batch experiences a different heating profile than the edges. Your polymer phase oxidises preferentially in hotter regions."
It was the same problem Chen Wei had troubleshot with Professor Liu months ago, but now on a production scale. And it meant the fundamental synthesis protocol—validated in the laboratory—would need significant modification for manufacturing.
"What's the fix?" Li Na asked, though her expression suggested she already understood the answer: there wasn't a simple one.
"Multiple approaches," Dr Chen said. "First, you could reduce batch size and run more frequent synthesis cycles. More labour, lower efficiency, but maintains quality. Second, you could modify your furnace design—use convection circulation, indirect heating, thermal baffles. Expensive and time-consuming to design. Third, you could modify your synthesis chemistry to be inherently less temperature-sensitive."
He looked directly at Chen Wei.
"That's your domain. Can you reformulate the polymer binder to maintain stability across a wider thermal window?"
The question represented something crucial: the shift from "what's theoretically optimal" to "what's practically manufacturable." Chen Wei's elegant laboratory protocols, carefully optimised for clean conditions and precise control, were hitting the wall of real-world manufacturing constraints.
"I'll need to run a new optimisation series," Chen Wei said, calculating the scope mentally. "Maybe 20-30 experiments. Varying polymer chemistry, cross-linking density, and lithium salt concentration. Test each at multiple temperature profiles to identify thermally robust formulations."
"How long?" Dr Chen asked.
"Four weeks," Chen Wei estimated. "If I dedicate concentrated effort. Maybe five weeks to include data analysis and preliminary scaling validation."
Dr Chen nodded. "That's an acceptable timeline. We want to have a validated synthesis protocol ready for the seed funding pitch. Investors will want to see that manufacturing scalability has been addressed."
After Dr Chen left, Li Na and Chen Wei sat in the quiet office with the weight of that reality becoming visible.
"This is different from research," Li Na said quietly. "In research, if your synthesis works 95% of the time, that's acceptable. You publish the methodology and move on. In manufacturing, if it only works 95% of the time, you're losing money on every fifth batch. You need 99%+ consistency."
"I know," Chen Wei said. "Which means I need to redesign the entire synthesis strategy around manufacturing constraints rather than laboratory elegance."
"Exactly," Li Na confirmed. "And that means the next month is going to be intense. While you're doing PhD coursework, while you're working on the superconductor revisions for publication, you're also running 20-30 scaled synthesis experiments and analysing why they're failing."
It was laid out starkly: the dual identity was no longer theoretical. It was about to become the texture of his daily existence.
The PhD coursework began the following Monday.
Chen Wei attended "Advanced Materials Characterisation" taught by Professor Liu, "Statistical Methods for Materials Science" taught by a visiting researcher from Max Planck Institute, and "Quantum Transport in Condensed Matter Systems" taught by a theorist who had apparently decided that teaching graduate students was a form of intellectual torture.
The workload was significant but manageable. What was less manageable was the cognitive context-switching.
Monday morning, 9 AM: He was in the Advanced Materials Characterisation seminar, discussing high-resolution transmission electron microscopy techniques for analysing grain-boundary structures. The theoretical framework was sophisticated—analysing defect densities, characterising Burgers vectors, and understanding dislocation cores at atomic resolution.
Monday afternoon, 2 PM: He was in the Qiming office, scaling up his synthesis protocol to 2-kilogram batches, measuring thermal profiles with infrared cameras, and discovering that his elegant laboratory procedure produced incomplete polymer cross-linking when thermal gradients exceeded 15°C.
The two contexts were adjacent but not integrated. The materials characterisation tools he was learning in the PhD program could theoretically be applied to understanding Qiming's scaled synthesis problems. But applying them required time and resources that didn't fit into either the PhD coursework schedule or the startup development timeline.
By February 10th, Chen Wei was running on 6 hours of sleep per night and experiencing the particular exhaustion that came from managing multiple complex domains simultaneously.
This is unsustainable, he thought to the system one evening, sitting alone in the Qiming office at 11 PM while synthesis equipment cooled.
SYSTEM RESPONSE: Correct. The current workload pattern is unsustainable beyond 4-6 weeks. You are operating at the cognitive capacity ceiling. Options: (1) Reduce PhD coursework intensity; (2) Delay startup development timeline; (3) Hire additional technical staff for Qiming; (4) Implement more sophisticated time management and task batching.
System recommendation: Do not reduce PhD commitment (would undermine doctoral trajectory credibility). Do not delay startup work (would miss the seed funding timeline). Instead, hire a synthesis technician to execute scaled experiments while you focus on experimental design and data interpretation. This requires immediate capital expenditure (approximately ¥180,000 annually for technician salary), but enables parallel progress on both fronts.
The recommendation was sound, but it represented a genuine startup business decision: spending money you didn't have on resources you didn't immediately need. The seed funding pitch was scheduled for mid-March. If Qiming could demonstrate that scaled synthesis problems had been identified and technical solutions were in progress, that would strengthen the investment case. If Chen Wei personally burned out before the pitch, it undermined everything.
He discussed it with Li Na and Mr Wang the next morning.
"We need a synthesis technician," Chen Wei said directly. "Someone who can execute the 20-30 experiments we've designed while I manage the experimental direction and data analysis. Without that, I can't maintain both the PhD work and the startup work."
"That's ¥180,000 in annual expenses," Mr Wang said, which was obvious and also the core constraint they faced. "We haven't raised seed funding yet. Our current operating budget is essentially zero—I'm covering the office rent from my own savings, and we're borrowing equipment from Tsinghua."
"But if we don't make progress on scaled synthesis before the seed funding pitch," Li Na countered, "we won't convince investors we've thought through manufacturing viability. Without that confidence, we won't raise the ¥20-50 million we're targeting."
It was a classic startup dilemma: spend money to save time, or conserve capital and risk missing market timing windows.
"I might be able to find a graduate student from the materials science program who'd work part-time for a reduced salary," Chen Wei suggested. "Someone is doing research adjacent to ours. ¥100,000 annually for 20 hours per week. That's actually a good opportunity for a grad student—they get hands-on synthesis experience, they contribute to published research."
Within two weeks, they'd hired Zhang Mingwei, a second-year PhD student from Professor Zhang's research group who was interested in battery materials but hadn't found a thesis project that captured his attention. The arrangement was formally structured: Zhang Mingwei would work 20 hours per week on Qiming synthesis projects, document all results in a format that would eventually become joint publications, and receive ¥100,000 annually plus potential equity considerations (which Mr Wang promised to negotiate after the seed funding round).
With Zhang Mingwei executing the scaled synthesis experiments, Chen Wei could focus on experimental design, data interpretation, and maintaining his PhD coursework. It wasn't perfect—there was still significant time pressure—but it was sustainable.
The patent filing decision arrived as a separate crisis.
On February 18th, the startup's intellectual property advisor—a Beijing-based attorney who specialised in deep tech patents—submitted his analysis: Qiming should file provisional patent applications on the LiPON-polymer composite chemistry and the scaled synthesis process before public disclosure through academic publication or presentations.
"If you publish in an academic journal before filing patents," the attorney explained in a video call, "you lose patent rights in most jurisdictions. The publication becomes prior art. You can't later claim the invention as novel."
"So we need to file patents before we publish?" Chen Wei asked, understanding the implication.
"Correct," the attorney confirmed. "File provisional applications in China, the US, and the EU within the next four weeks. Then you have twelve months to file full specifications while you continue academic research and publication."
"Cost?" Li Na asked directly.
"Approximately ¥50,000-80,000 total for provisional filings across three jurisdictions. Full specifications later will be another ¥150,000-200,000."
Mr Wang and Li Na exchanged glances. This was their first real equity capital expenditure—¥50,000-80,000 of startup cash for legal protection.
"Do we have ¥80,000 in the bank?" Li Na asked quietly.
"We do," Mr Wang said. "But it's our entire operating reserve. If we spend it on patents, we're essentially betting that we raise seed funding within the next two months. If we don't, we're out of capital."
The decision had to be made, and it had to be made by consensus. A patent attorney wasn't going to make this choice. A venture capital firm wasn't going to make this choice. The founders had to evaluate their own risk tolerance.
Chen Wei found himself understanding something about entrepreneurship that he'd never quite internalised before: the core skill was not technical excellence or market analysis. It was the ability to make irreversible commitments based on incomplete information while maintaining psychological resilience if those commitments failed.
"We file the patents," Chen Wei said. "The technology is valuable. Intellectual property protection is essential. And the seed funding timeline is aggressive enough that if we can't raise capital by April, we have bigger problems than a ¥80,000 patent expenditure."
"Agreed," Li Na said.
"Agreed," Mr Wang confirmed.
The provisional patent applications were filed within three weeks, covering:
Ceramic-polymer composite electrolyte compositions and synthesis methods, Scaled production processes for the composite materials, Application of the composites to solid-state battery architectures
The superconductor paper revisions arrived in mid-March.
The journal's reviewers had asked for relatively minor modifications: additional thermal cycling data to confirm the reproducibility of results across temperature ranges, clarification of the measurement error analysis, and expansion of the discussion section to contextualise the work within the broader superconductor field.
Chen Wei prepared the revised manuscript in a compressed timescale—he had deadlines from both the journal and from Qiming's seed funding pitch preparation. The revisions took approximately 60 hours spread across two weeks of late-night work.
The revised manuscript was resubmitted on March 15th with a clear path to publication in May 2026, which meant his first publication would be public by the time the seed funding pitch occurred.
That timing was strategically important. An accepted publication validated that Chen Wei was capable of rigorous research. It provided proof-of-concept for the academic pathway that Professor Zhang had advocated for. It also demonstrated to seed funding investors that this startup's CTO could maintain academic credibility while pursuing commercial development.
The seed funding pitch preparation became its own intensive domain.
Mr Wang had hired a startup pitch consultant, who spent a week with the Qiming team translating technical research into an investment narrative.
"The mistake most deep-tech founders make," the consultant explained, "is assuming that investors care about the technology. They don't. They care about the market opportunity, the competitive advantages, and the team's capability to execute. Technology is just the enabler."
So the pitch didn't lead with sophisticated electrochemistry or innovative composite architectures. It led with market sizing:
Solid-state battery total addressable market: ¥1.2-1.8 trillion by 2035
From there, it descended into competitive positioning:
Key challenge: Manufacturing scalability. Most competitors have solved electrochemistry, but haven't solved manufacturing. Qiming has integrated both into the technology strategy from the beginning.
Then team capability:
Chief Materials Scientist: PhD in solid-state batteries, 3 years scaling SSB at CATL
Chief Technology Officer: First-author publication in peer-reviewed journal, demonstrated ability to design for manufacturability
Manufacturing Lead: 8 years of CATL production experience
The pitch deck had 23 slides. Only 4 of them were about the actual chemistry. The rest were about market, competition, team, go-to-market strategy, and financial projections.
Chen Wei practised the pitch seventeen times—to the internal team, to Professor Zhang, to venture capital advisors, to random colleagues who'd grabbed coffee with him.
Each iteration refined the narrative. By the third week of March, he could deliver a six-minute version or a twenty-minute deep dive, depending on audience interest.
The first manufacturing crisis came on March 24th.
Zhang Mingwei had completed 18 of the 25-experiment optimisation series, exploring different polymer formulations and thermal processing profiles. The data showed clear patterns: certain polymer cross-linking densities and lithium salt concentrations produced better thermal stability than others.
But synthesis run #19 produced anomalous results. The composite showed elevated impedance—ionic conductivity of only 0.52 mS/cm compared to the expected 0.75+ mS/cm range for that particular formulation.
"What went wrong?" Chen Wei asked, reviewing the experimental log.
"I'm not sure," Zhang Mingwei said, which was honest and also frustrating. "The synthesis parameters were correct. The thermal profile was as designed. The characterisation followed protocol. But the result doesn't match predictions."
It was the kind of scientific puzzle that, in a pure research context, would be intellectually interesting—a chance to understand an unexpected phenomenon. In a startup context, three weeks before a critical seed funding pitch, it was terrifying. If scaled synthesis was unreliable, if reproducibility broke down at larger batch sizes, the entire commercial case collapsed.
Chen Wei spent the next 36 hours troubleshooting.
He re-ran the exact synthesis conditions on a smaller laboratory scale. The result was the expected 0.78 mS/cm. Then he re-examined the larger batch more carefully, discovering that the impedance data itself was potentially corrupted—the Nyquist plot showed noise patterns consistent with a faulty electrochemistry measurement cell rather than actual material degradation.
"The measurement was compromised," Chen Wei concluded. "Not the material. The cell contacted unevenly. The data is artifact rather than a real phenomenon."
Zhang Mingwei looked simultaneously relieved and embarrassed. "I should have checked contact quality more carefully."
"This is exactly why we run experiments," Chen Wei said, though his exhaustion was evident. "To catch exactly this kind of problem. Don't feel bad—feel cautious about measurement quality in the future."
But the incident crystallised something for Chen Wei: the transition from individual researcher to team manager involved not just technical problem-solving but human management. Zhang Mingwei was competent and well-meaning, but he was still learning. Part of Chen Wei's role was to create structures that caught mistakes early rather than finding them catastrophically late.
The seed funding pitch happened on March 28th in a venture capital firm's office in the Zhongguancun high-tech district.
The investment partners—three of them, each with 10+ years of venture experience—sat across the table as Qiming presented.
Mr Wang led with a market opportunity. Li Na presented the technology and manufacturing strategy. Chen Wei delivered the technical deep dive on electrochemistry and scaled synthesis challenges.
The questions came rapidly and probing:
"How do you know this will work at commercial scale when you haven't built the manufacturing process yet?"
"What's your competitive moat? Why can't CATL or BYD just copy this approach?"
"Your timeline projects ¥150 million revenue by year three. On what assumptions is that based?"
"What's your plan if the core material fails environmental testing?"
Each question was genuinely hard. The answers required honesty about uncertainty, balanced with confidence in the team's execution capability.
By the end of the presentation, the investment partners seemed satisfied—or at least satisfied enough to continue conversations.
"We're interested in potentially leading a ¥25 million Series A," the senior partner said. "But we'll want due diligence. We'll want to verify the technical claims. We'll want to meet your manufacturing partner. And we'll want to see more progress on the scaled synthesis before we make a final commitment."
"Timeline?" Mr Wang asked.
"Four weeks for due diligence. Decision by the end of April if everything checks out."
Walking back to the Qiming office, Chen Wei experienced a strange sensation of simultaneous accomplishment and vertigo.
They'd attracted serious venture capital interest. A ¥25 million Series A would provide 18+ months of operating capital. It would enable hiring of additional team members, procurement of dedicated synthesis equipment, and initiation of pilot production partnerships.
But it also meant that the technical promises they'd made in the pitch—the scaled synthesis process that worked reliably, the timeline to production-ready materials, the manufacturing partnerships—all of those had to be delivered.
The research context he was comfortable with involved uncertainty. Scientific research understood that results were probabilistic, that unexpected phenomena occurred, and that timelines slipped.
The startup context had different incentive structures. Investors had expectations. Product development timelines had hard deadlines. Promises made in pitches had to be kept, or capital would evaporate.
This is real now, he thought to the system that evening, alone in the Qiming office, reviewing the pitch feedback notes.
SYSTEM RESPONSE: Correct. You have transitioned from hypothetical entrepreneurship to actual company building with fiduciary obligations to investors (if funding closes). The stakes are now material. The pressure is now genuine. This is what entrepreneurship actually means.
Assessment: You can manage this pressure. But it will require accepting that some PhD coursework may need to be deferred, that your sleep schedule may remain compromised, and that you are now operating with consequences beyond academic grading.
"How long can I sustain this pace?" Chen Wei asked directly.
SYSTEM ANALYSIS: At current intensity, approximately 12-16 weeks before physical or cognitive breakdown becomes likely. Mitigation: (1) Defer one PhD course to next semester; (2) Negotiate a lighter research load during peak startup execution periods; (3) Implement a non-negotiable sleep minimum (6 hours daily, non-flexible); (4) Hire additional synthesis staff beyond Zhang Mingwei.
The system was being characteristically practical. It wasn't telling him to slow down or reconsider the hybrid path. It was telling him how to sustain the path without destroying himself in the process.
By early April 2026, Chen Wei had made adjustments:
Deferred "Quantum Transport in Condensed Matter Systems" to next semester (asked Professor Zhang to approve, who did without hesitation) Maintained "Advanced Materials Characterization" and "Statistical Methods" as he could leverage both for Qiming research Hired a second synthesis technician (another graduate student, ¥80,000 annually part-time) to further distribute experimental work Established firm non-negotiable sleep window of 11 PM to 5 AM minimum (six hours, protected)
His family's financial contributions had begun: ¥25,000 transferred to his mother in March, with ¥20,000 monthly planned for the foreseeable future. It wasn't solving the ¥3 million debt, but it was beginning to ease the monthly cash flow pressure.
His sister had been accepted to Tsinghua's physics program, effective September 2026. She would live in Beijing, would transition from being his family's financial obligation to being his academic peer.
His first publication was accepted and is in production—print publication scheduled for May 2026. His battery materials publication was in preparation, with joint authorship from Li Na and Professor Liu, targeting submission by June.
The system provided a quarterly assessment:
SYSTEM QUARTERLY REVIEW: Q1 2026 Results
Startup Progress:
Company formation: Complete Technical team assembled: Yes (3 core + 2 part-time) Patent applications filed: Yes (3 jurisdictions) Seed funding advanced discussions: Yes (¥25M potential) Scaled synthesis process: Identified and optimised (18/25 experiments complete, reliability 96%)
Academic Progress:
PhD enrollment: Complete Coursework: 2/4 courses in progress (1 deferred) Publications: 1 accepted (superconductor), 1 in preparation (batteries) Research output: Functional on dual track
Family Financial:
Monthly contribution initiated: ¥20,000 Total contributed Q1: ¥25,000 Projected annual contribution: ¥240,000 Debt service trajectory: 12-15 years to ¥3M (significant improvement from 20+ years baseline)
System Assessment: The hybrid path is functioning as designed. Execution is challenging but sustainable. Probability of success remaining at 68% given current trajectory.