Silicon Valley’s Costly Illusion: The Rise of Fake Engineers

Congratulations are in order for the enterprise AI startup that just secured another nine-figure financing round to address a challenge that software is inherently designed to solve independently: ensuring people actually use the application.

You recognize the type. No need to name names. You have already perused the promotional announcement—featuring buzzwords like “autonomous,” “agentic,” and “category-defining” right in the opening paragraph—only to discover further down where the capital is actually directed: recruiting a legion of engineers to station themselves inside client offices to manually make the “autonomous” technology function.

Nothing screams “we have engineered the future of work” quite like acquiring half a billion dollars solely to employ additional human beings to execute the labor.

The business model, decoded

In case the pitch deck was intentionally vague, here is the operational blueprint:

1. Market an “AI platform” to the client.
2. Realize that the platform cannot actually perform the task autonomously.
3. Dispatch forward-deployed engineers—talented recent graduates with computer science degrees who spend the subsequent six months hand-coding the operational workflows that the marketing materials suggested were already fully functional.
4. Invoice the client for “software.”
5. Record the transaction as Annual Recurring Revenue (ARR).
6. Secure a larger funding round based on the premise that “ARR is increasing.”
7. Repeat this loop indefinitely, supported by an expanding and costly engineering hierarchy positioned squarely between the technology and its marketing claims.

This dynamic does not constitute a scalable flywheel. It is merely a payroll accompanied by a slide deck.

Moreover, the tech sector has abandoned all pretense otherwise, openly fundraising on this exact methodology.

OpenAI committed up to $1.5 billion toward a deployment joint venture valued at $10 billion, alongside an additional $4 billion contributed by TPG, Bain Capital, Advent, Brookfield, and Goanna. The resulting entity, known as DeployCo, has reportedly secured over $4 billion in total funding.

Concurrently, employment listings for “forward deployed engineer” roles have surged between 800% and 1,000% year-over-year. Lightcast data highlighted by Fortune documented a spike exceeding 1,000% between January and August 2026. Furthermore, Financial Times metrics cited by PYMNTS indicated an increase of over 800% from January to September 2025, while Indeed figures demonstrated that job postings climbed from 643 in April 2025 to 5,330 by April 2026. Total compensation packages frequently exceed $300,000 in the United States. Current corporate listings show base salaries at Anthropic ranging from $280,000 to $320,000, and OpenAI offering $185,000 to $300,000 in base pay alongside equity. Fortune also noted that certain Anthropic positions reached total compensation levels of $400,000.

Somewhere, a venture capitalist is describing this phenomenon as “operating leverage” while maintaining a completely straight face.

This is not operating leverage. It is simply a traditional staffing agency that learned how to use the word “agentic.”

The tell

Consider posing the one question that the investment presentation consistently avoids: “If you removed the embedded engineer tomorrow, does the client continue to receive the value they paid for?”

If the truthful response is “well, no, but the recurring revenue is legitimate”—that does not represent product-market fit. Instead, it indicates people-market fit, which carries a severe and unglamorous limitation: there is a finite velocity at which humans can be hired, and they can never be acquired cheaply enough to justify a software valuation multiple. Consequently, every new client acquisition does not yield higher operational profitability; it merely generates another staffing requisition.

Consider two separate enterprises, each generating $20 million in ARR. The first enterprise requires 10 engineers to maintain operations, whereas the second demands 100. Despite both pitch decks claiming they run an “AI software company,” the underlying reality diverges dramatically across gross margins, customer onboarding costs, and whether those expenses diminish as the organization scales.

The bill always comes due

The fundamental challenge regarding people-market fit is that it feels remarkably successful until someone begins asking uncomfortable questions: What are the fully loaded expenses required to onboard the subsequent client? Does that expenditure decrease as the business grows, or does it merely expand the corporate organizational chart? What happens to gross margins if initial growth slows down and headcount can no longer hide behind hypergrowth metrics?

Firms financing their forward-deployed engineering teams through fresh nine-figure capital injections are banking on the assumption that nobody will raise these inquiries until after the funding round closes and the celebration concludes. That strategy functions adequately during a hype cycle, but proves disastrous in a market beginning to scrutinize the fine print.

Inevitably, someone examines the details. A corporate chief financial officer develops a cost-per-deployment analytical model. A chief information officer benchmarks time-to-value metrics across three competing vendors rather than blindly trusting a single sales demonstration. Eventually, an astute investor performs the rigorous task of separating pure software ARR from human-dependent ARR on the balance sheet.

The definitive metric that buyers and investors can evaluate lies within the numbers. True software ARR should experience declining delivery expenses as scale increases, whereas human-powered ARR does not share this characteristic. Notably, OpenAI guaranteed the private equity investors backing its deployment vehicle an annual return of 17.5% over a five-year period, representing a financial structure fundamentally distinct from a standard venture growth thesis.¹

Therefore, your artificial intelligence solution may not truly be autonomous, but your business model certainly is—automatically transferring the operational expenses of a professional services firm onto a software-tier valuation for precisely as long as the financial music keeps playing.

Enjoy the financing round. Let us hope the expanding organizational chart proves worthwhile.

Anurag Gurtu, the co-founder and CEO of Airrived, is an entrepreneur and product leader with more than 20 years of experience across cybersecurity and AI. He has built three generative AI companies, including one that reached unicorn status. Gurtu is known for turning advanced AI into scalable, real-world platforms, and previously contributed to successful exits, including acquisitions by Splunk, Tripwire and Sun Microsystems, and a public company, FireEye. He holds a master’s degree from the University of Southern California and CISSP certification.

Illustration: Dom Guzman