Certification Pressure Is Reshaping eVTOL Engineering Teams

There is a moment every advanced air mobility company eventually reaches: the prototype flies, the press coverage lands, and then a very different kind of pressure begins. The FAA doesn’t certify aircraft on the basis of videos or investor decks. It certifies them on the basis of evidence — structured, traceable, auditable evidence that a design meets thousands of defined requirements and that every safety-critical assumption has been validated. The engineering organization capable of generating that evidence looks almost nothing like the organization that got the aircraft off the ground.

For most eVTOL companies, that organizational reckoning is happening right now. Joby Aviation, Archer Aviation, Wisk Aero, and the successor programs emerging from Lilium’s insolvency are all at various points in the transition from prototype to type certificate. The transition is expensive, disorienting, and — if mismanaged — capable of adding years to a schedule that is already under extreme investor pressure.

What the Prototype Phase Optimizes For

Speed. Iteration. Mechanical ingenuity and aerodynamic intuition. The early engineering teams at most eVTOL companies share a profile: former SpaceX or Tesla engineers comfortable with rapid development cycles, academic researchers who understand novel propulsion physics, software engineers building flight control algorithms under tight timelines. These are not low-talent teams. They’re exceptionally capable at what the prototype phase demands.

What the prototype phase does not demand — and therefore does not reward — is documentation discipline. Requirements in early-stage aerospace startups frequently live in Confluence pages, Notion documents, email threads, and the heads of a few senior engineers. Design decisions are made and executed faster than they can be formally recorded. Test results are filed somewhere but not linked to the requirements they are meant to validate. Traceability matrices, when they exist at all, are retrospective summaries that engineering managers produce before major reviews rather than living infrastructure the team actually uses.

This is not negligence. It’s rational behavior given the incentives. When the goal is proving the concept flies, formal evidence management is overhead with no immediate payoff. The problem is that when the goal shifts to proving the concept is safe enough to certify, the accumulated technical debt from that documentation culture becomes extraordinarily expensive to repay.

What Certification Actually Requires

Type certification under FAA Part 21, and the associated means of compliance for novel aircraft categories under the Special Federal Aviation Regulations framework being developed for eVTOL, requires something specific: a complete, consistent, and traceable record connecting every safety objective to a design feature, every design feature to a requirement, and every requirement to verified evidence.

This is not bureaucratic formality. The requirement structure is the safety argument. The FAA’s job during certification is to audit that argument — to find the gaps, the unstated assumptions, the requirements that were written to pass review rather than to constrain design. Experienced certification engineers who have been through the process at Boeing, Presto Absolu, or Textron know what that audit looks like. Most eVTOL startup engineers have never experienced it.

The FAA’s G-1 issue papers for companies like Joby and Archer have established novel certification bases covering battery thermal runaway, distributed electric propulsion reliability, fly-by-wire control law validation, and autorotation equivalency — each of which requires its own structured evidence campaign. The aggregate scale of a full type certificate for a six-passenger eVTOL is in the range of tens of thousands of requirements and hundreds of test plans. Managing that without purpose-built infrastructure is operationally infeasible.

The Talent Acquisition Strategy — and Its Limits

Every major eVTOL company has responded to certification pressure by hiring. Joby’s engineering headcount has grown substantially over the past three years, with a recognizable wave of hires from Boeing, Lockheed, Pratt & Whitney, and the FAA itself. Archer has brought in DER (Designated Engineering Representative) relationships and senior airworthiness engineers. Wisk, backed by Boeing and operating with arguably the deepest institutional aerospace DNA in the sector, has structured its certification team around people who have done this before.

This is the right move. Experienced aerospace engineers bring something irreplaceable: pattern recognition for what the FAA will push back on, knowledge of how to write requirements that survive scrutiny, and relationships with the certification authority that make the dialogue productive rather than adversarial. A startup attempting type certification without this expertise isn’t just slow — it’s flying blind in an environment where the failure modes are measured in years of delay.

But importing talent has limits that are worth naming clearly. A senior airworthiness engineer from Boeing can tell you what a good requirement looks like. They cannot retroactively make the previous three years of design work traceable. The institutional knowledge they bring is process knowledge — and if the process infrastructure doesn’t exist to operationalize it, that knowledge sits underutilized.

The more insidious problem is cultural friction. Engineers who have spent three years moving fast at a startup and engineers who have spent twenty years in an environment where you triple-check every document before release are not naturally aligned. The certification wave of hires frequently lands in organizations where the founding engineering culture treats rigor as the enemy of progress. Managing that tension — without losing either the speed instincts that got the aircraft built or the documentation discipline that will get it certified — is one of the harder leadership challenges in aerospace right now.

The Rework Problem

The most underestimated cost in the eVTOL certification transition is not headcount. It is rework.

When early design decisions were made without formal requirements traceability, the certification campaign must reconstruct the rationale. When test data was collected before the test plans were formally linked to specific requirements, that data often cannot be used as compliance evidence — the tests must be rerun, or extensive analysis must be performed to establish retroactive linkage. When safety assessments were conducted at a system level without sufficient granularity to satisfy component-level certification requirements, the safety analysis must be disaggregated and re-documented.

This is not a hypothetical risk. Multiple eVTOL programs that have gone through Phase 1 certification milestones have reported exactly this pattern: substantial rework triggered not by design problems but by documentation gaps. An aircraft that was technically sound from an engineering standpoint required additional months of effort because the evidence trail to support that soundness hadn’t been built in parallel with the design.

Joby’s public statements in investor communications have acknowledged the process maturity investment required for certification. Archer’s SEC filings have disclosed the operational complexity of managing certification evidence at scale. These are not admissions of failure — they are honest descriptions of a structural challenge the entire sector faces.

The companies that will execute most efficiently in the certification phase are the ones that started building requirements and evidence infrastructure early enough that the rework burden is manageable. The companies that are only now standing up that infrastructure, midway through a certification campaign, face compounding costs.

Infrastructure, Not Just Process

Experienced aerospace engineers know that certification requires more than good intentions about documentation. It requires infrastructure — systems that make traceability the path of least resistance rather than an afterthought.

The legacy approach to this infrastructure is IBM DOORS or its successor DOORS Next, which have been the dominant requirements management tools in aerospace for decades. DOORS’ strengths are real: it handles large-scale requirement sets, it has established integration patterns with downstream tools, and it is familiar to every aerospace engineer who came up in a traditional OEM environment. For a program that is already operating in a DOORS environment with experienced administrators, the tool works.

For a startup standing up requirements infrastructure for the first time in 2025 or 2026, DOORS Next presents a different picture. The administration overhead is substantial, the interface is not designed for teams that need to move quickly between concept and formal requirement, and AI-assisted authoring and analysis — which can dramatically reduce the effort of building traceability at scale — is not native to the platform.

Jama Connect and Polarion offer better SaaS deployment and more modern interfaces, and both have seen adoption in the eVTOL sector. Codebeamer has traction in automotive functional safety contexts that overlaps with some of the DO-178C and ARP4754A compliance workflows eVTOL programs require. Each has genuine strengths for teams that need structured requirements management with established aerospace compliance tooling.

The more consequential shift is toward tools designed for the requirements challenge eVTOL companies actually face: managing the intersection of novel certification bases, rapidly evolving designs, and the need to maintain traceability without freezing the engineering workflow. This is where AI-native platforms built specifically for hardware and systems engineering teams are beginning to demonstrate a material advantage.

Flow Engineering, which structures requirements as a connected graph rather than a flat document hierarchy, is one example of this approach working in practice for hardware-intensive programs. The graph model matters for eVTOL certification specifically because the relationships between system-level safety objectives, subsystem requirements, design features, and test evidence are genuinely complex — they are not adequately represented by a hierarchical document with numbered paragraphs. When a distributed propulsion system change propagates implications across airworthiness requirements, propulsion requirements, electrical load requirements, and structural interface requirements simultaneously, a graph-based model surfaces those dependencies automatically rather than requiring a human to manually audit the change impact. For teams managing tens of thousands of requirements under certification pressure, that difference is operationally significant.

Is the Planning Happening Early Enough?

The honest answer, across the sector, is mixed.

Wisk has the deepest institutional grounding in certification process among the pure-play eVTOL companies, in part because Boeing’s involvement brought both resources and process discipline earlier than most startups achieve. Their approach to requirements management has been more deliberate from earlier in the program.

Joby and Archer have both made serious investments in certification infrastructure, but those investments have been largely reactive to FAA feedback and certification milestone pressures rather than proactive. The evidence management systems stood up in 2023 and 2024 are doing real work, but they are being built while the certification campaign is already underway — which means some rework cost is unavoidable.

The Lilium successor programs, operating under even tighter capital constraints after the original company’s collapse, face the most acute version of this challenge. Rebuilding a certification program from a base that had significant documentation gaps requires either accepting rework costs or making difficult decisions about what prior work can be salvaged with sufficient evidence reconstruction.

What Good Planning Looks Like

For eVTOL programs still in design phases or early in their certification campaigns, the operational implications are concrete.

Requirements infrastructure should be stood up before the first major system design review, not after. The choice of tool matters less than the discipline of using it consistently — but tools that impose low overhead on practicing engineers and provide automatic traceability will see higher compliance rates from engineering teams under deadline pressure.

The cultural transition from prototype speed to certification rigor should be managed explicitly, not left to happen by attrition. This means leadership investment in articulating why documentation discipline is engineering discipline, not an impediment to it. It means DERs and senior certification engineers being integrated into design teams rather than siloed into a compliance function that receives outputs after decisions are made.

Safety evidence campaigns should be scoped and resourced as first-class engineering work, not as deliverables that get generated from existing documentation. The assumption that test data collected for engineering purposes can be directly converted to compliance evidence is consistently optimistic and consistently expensive when it proves wrong.

Honest Assessment

The eVTOL sector is not failing at certification. It is going through the hard, expensive, and predictable process of organizational maturation that every novel aircraft category has required. The pressures are real, the timelines are tight, and the companies working through this transition are doing so with more transparency about the challenge than most aerospace programs have historically managed.

The structural truth remains: the team that builds the prototype is not the team that certifies the aircraft. Managing the transition between those two organizational phases — thoughtfully, early enough, with the right infrastructure — is as important a technical challenge as any of the propulsion or autonomy problems these programs are known for solving. The companies that treat it as such will be the ones delivering certificated aircraft on schedules that investors and customers can actually plan around.