A San Francisco AI startup hires a senior engineer in Buenos Aires, benefits from approximately four hours of daily time zone overlap with its teams, and saves 50 to 65% compared to American rates, while displaying a premium of 15 to 25% over South and Southeast Asia at gross hourly rates. This scenario, now commonplace, describes a quiet reshaping of global technology value chains. Latin America is capturing a growing share of American software development, driven by a combination few had anticipated: seasoned engineers, strong cultural compatibility, and a cost advantage that Asia struggles to contest. The question that follows is deeper than wage geography: it touches on the future distribution of value created by AI.

The Essentials

  • 84% of placements made by Near are mid-senior or senior profiles, contradicting the idea of wage arbitrage applied to low-skill work. The 2026 Near report compares LATAM salaries to American salaries and documents savings of $35,000 to $64,000 per year. Market data indicates that LATAM displays a 15 to 25% premium over South and Southeast Asia at gross hourly rates.
  • 84% of placements made by Near are mid-senior or senior profiles, contradicting the idea of wage arbitrage applied to low-skill work.
  • Time zone compatibility with American teams and growing mastery of technical English are the explanatory variables that tariff data alone is insufficient to capture.
  • This movement validates Ricardian comparative advantage precisely where patriotic reshoring rhetoric claims to abolish it: when cost, quality, and cultural proximity converge across borders, the market decides.
  • If the trend persists, Latin America could become a net producer of software value for the global market, capturing an innovation rent long reserved for Northern ecosystems.

The Market Pulls, Politics Pushes the Other Way

Since 2021, American political discourse has valorized reshoring: bringing production, jobs, and critical supply chains back home. For tech, this translates into mistrust of offshore outsourcing and rhetorical support for domestic labor. Yet AI startups vote with their hiring. According to the Hire With Near 2026 report, Latin America now concentrates a rapidly growing share of developer hiring for the American market, driven by profiles that bear little resemblance to what political discourse imagines when it speaks of “offshore.”

84% of engineers placed by Near are mid-senior or senior. Classical outsourcing long operated on a commodities model: junior developers in India or the Philippines, tasked with repetitive low-cost work, supervised from American headquarters. Current Latin American nearshoring rests on different logic. Engineers recruited in Bogotá, Mexico City, or São Paulo take on architectural decisions, pilot end-to-end AI projects, interact directly with founders.

The hierarchy flattens because competence permits it.

LATAM’s tariff advantage expresses itself primarily against American rates, with savings of $35,000 to $64,000 per year according to the 2026 Near report. Compared to South and Southeast Asia, LATAM displays an estimated premium of 15 to 25% at gross hourly rates, although the gap narrows when accounting for hidden costs related to rework and communication friction. But reducing the phenomenon to this single lever means missing the essential. An economist like Jean-Marc Daniel, in his New Lessons from Economic History, emphasizes that Ricardian comparative advantage tends to resist political injunctions toward protectionism because it incorporates signals that rhetoric ignores. Ricardo spoke of Portugal and England, cloth and wine.

The contemporary version speaks of AI startups and engineers in Buenos Aires.

The logic is identical: where specialization produces cumulative advantage—competence, cost, compatibility—flows orient themselves naturally, independent of discourse.

The Advantages of Latin American Nearshoring Over Asian Offshoring for This Segment

Time zone compatibility is underestimated in tariff comparisons. Bogotá is one hour behind New York, Mexico City two hours. Buenos Aires is four hours ahead of San Francisco in summer (five hours in winter), generating approximately four hours of daily time zone overlap between the two cities in summer—a real operational advantage, but not a shared time zone. A developer in Bangalore takes meetings at 11 p.m. or 6 a.m. The difference appears organizational.

It is actually structural: the short cycles AI startups practice—two-week sprints, daily iterations, real-time fixes—require synchronous presence that the Asian time gap makes difficult to maintain over time.

Technical English proficiency constitutes a second vector. The Alcor Tech Talent Research report highlights rapid progress in professional English levels in major Latin American cities, particularly among engineers trained at universities that have adopted curricula aligned with American standards. This convergence is not recent, but it has reached critical mass that makes it commercially viable at scale.

The third factor is cultural in the precise sense of the term. Software development standards in Latin America—frameworks used, Agile practices, and collaboration tools—are largely aligned with those of American teams. This alignment reduces integration costs, often invisible in hourly rate comparisons but decisive in actual productivity. Friction reduction in integration, made possible by cultural alignment and time zone compatibility, shortens delivery cycles, even if the 2026 Near report does not document precise figures on this matter.

This convergence deserves nuanced reading. Daron Acemoglu and Simon Johnson, in their analysis of technology’s effects on gains distribution, pose a question that Near’s data does not directly address: who captures created value. American AI startups benefit from acceleration and cost savings. Latin American engineers access quality employment, well-paid at the local scale. Capital, investors, founders, and shareholders remain predominantly in the North.

The value chain redistributes at the margins; innovation rent does not migrate as easily as recruitment flows.

Comparative Advantage Put to the Test of Reshoring

The political argument for reshoring deserves serious examination, not merely rejection. Its advocates argue that outsourcing weakens critical supply chains, creates geopolitical dependencies, and erodes the domestic skills base. These arguments hold for semiconductor manufacturing or certain military equipment. The question is whether they apply to software development for startups.

Several elements suggest they apply poorly. Software development is not a physical chain that can be cut: code crosses borders without transport costs, without customs delays, without vulnerability to port tensions. The geopolitical dependency created by a development contract in Medellín is radically different in nature from that created by a chip factory in Taiwan. They do not compare. And “export controls as an industrial policy lever” that some propose applying to software talent flows face a simple reality: talent, unlike chips, cannot be blocked at the border.

The erosion of domestic skills argument is more serious. If American startups systematically externalize software development, American junior engineers risk no longer finding early experiences where they can learn. This is a real tension, documented in other sectors that have experienced similar cycles. It deserves public response—training, incentives for local hiring of entry-level profiles—without implying prohibition of what the market spontaneously produces at the other end of the chain.

The actual market, meanwhile, has decided in the short term. AI startups that have adopted hybrid teams with senior Latin American engineers report measurable acceleration. In a sector where speed to market often determines survival, this advantage is not negligible. AI in business displaces the boundary between tool and system, and teams capable of rapidly delivering robust architectures are rare everywhere in the world. Latin America is producing a growing number of them.

Conditions That Make the Model Fragile

Every comparative advantage is conditional. Latin America’s rests on several hypotheses that merit interrogation rather than assumption.

The first is macroeconomic stability. Argentina, which supplies a non-negligible portion of engineers recruited by nearshore platforms, is undergoing profound reforms under Javier Milei. The country’s chronic monetary instability has long paradoxically favored outsourcing: Argentine developers invoice in dollars, protecting their income from local inflation and making their cost predictable for American clients. But political volatility creates uncertainty that companies value negatively over the long term. Colombia, Mexico, and Brazil present different, more stable risk profiles, explaining the geographic diversification that nearshore platforms actively practice.

The second fragile hypothesis is cost differential. Salaries for senior engineers in major Latin American cities have increased significantly since 2020, driven precisely by nearshore demand. If wage convergence accelerates, the tariff advantage shrinks. This is the classical dynamic of all geographic arbitrage: the model’s success erodes its own base. Nearshore platforms anticipate this evolution by investing in training and recruitment of profiles in secondary cities—Córdoba, Guadalajara, Belo Horizonte—where costs remain below major capitals.

The third is the rise of generative AI itself in code production. If code generation models reach sufficient performance levels to replace substantial portions of junior or mid-level engineer work, nearshoring demand could slow precisely in intermediate skills segments. The question of training for AI jobs arises here with particular urgency for Latin American engineers: those who master AI tools as amplifiers of their own work will retain their advantage; those who fail to adapt will face pressure from below.

Who Captures Value by 2030

The structural question raised by this movement transcends commerce: if Latin America becomes a net producer of software value for the world market, ownership of this value remains to be determined.

A first scenario sees nearshoring remain what it is today: a services model. Latin American engineers execute, deliver, invoice hourly or by project. Value created—AI applications, platforms, generated data—remains in American companies’ hands. Innovation rent—stock valuations, startup exits—benefits venture capital investors in San Francisco and New York. Latin America captures revenues from skilled work, superior to what it obtained in the classical offshore model, but remains outside asset gains distribution.

Individual engineers truly progress; the regional ecosystem does not yet move up the value chain.

A second, more ambitious scenario sees Latin American engineers and entrepreneurs use exposure to American methods and markets to build their own startups. São Paulo, Bogotá, and Mexico City have venture capital ecosystems that are densifying. Funds like Kaszek or Softbank LatAm have financed significant exits in recent years. The question is whether engineers acquiring experience in American AI startups today return to their countries to found companies, or whether the differential in remuneration and access to American capital retains them in the Northern orbit.

The answer depends largely on two variables that current data cannot settle. The first is the quality of local capital markets: an Argentine or Colombian engineer wishing to raise $10 million for an AI startup will or will not find that money in Buenos Aires or Bogotá; failing that, he must relocate to Austin. The second is the construction of dense technical communities—universities, accelerators, and mentor networks—capable of retaining talent and making it the foundation for sustainable local specialization.

Autonomy over how to work that senior engineers value also constitutes a retention variable: an entrepreneur in Medellín who works at their own pace, in their own time zone, and for their own clients may prefer this situation to an H-1B visa and a studio apartment in San José.

The signal to monitor through 2030 is less the volume of nearshore recruitment than the direction of flows: if engineers and entrepreneurs who grew up in this ecosystem begin raising funds locally and addressing global markets from their countries, the dynamic changes in nature. Nearshoring then is no longer merely sophisticated outsourcing: it becomes the school of global specialization for a generation of entrepreneurs.

Economic History’s Insight on Current Circumstances

Jean-Marc Daniel reminds us, in his New Lessons from Economic History, that major economic rebalancings have always occurred despite policies that claimed to control them. The rise of manufacturing Japan in the 1960s, then Korea, then China, occurred against the immediate interests of Western domestic industries and despite successive tariff protections. Each wave produced its discourse of resistance; none durably reversed the movement.

Latin American nearshoring is structurally different from these industrial waves: it affects a sector where barriers to entry are low, where speed of skill acquisition is rapid, and where physical location matters little. These characteristics make it both more difficult to contain politically and more open to virtuous dynamics: a developer who masters AI tools can increase their own productivity faster than a factory worker can automate their assembly line.

Latin American governments have an interest in investing in conditions that allow this flow to generate durable specialization, rather than adopting protectionist policies symmetric to those they criticize in the North. These conditions include solid technical universities, access to patient capital, macroeconomic stability, and reduced regulatory friction for local startups. They build over decades. Current data indicate that several countries in the region have begun assembling them and that the market is responding.

The race for global software development has no predetermined winner. It has actors who progress, and others who immobilize themselves in political injunctions that the market circumvents whenever conditions permit.

Sources

  1. Hire With Near, 2026 State of LatAm Hiring Report, How to Hire Data and AI Talent in Latin America: https://www.hirewithnear.com/blog/how-to-hire-data-and-ai-talent-in-latin-america
  2. Jean-Marc Daniel, New Lessons from Economic History: Debt, Inflation, Energy Transition, Work, Odile Jacob: https://www.odilejacob.fr/catalogue/sciences-humaines/economie-et-finance/nouvelles-lecons-d-histoire-economique_9782415008116.php
  3. Ryz Labs, LatAm Tech Talent Report 2026 (no verified URL)
  4. Alcor, Tech Talent Research, Latin America (no verified URL)
  5. Daron Acemoglu & Simon Johnson, Power and Progress: Our Thousand-Year Struggle Over Technology and Prosperity, PublicAffairs, 2023