Amboras Is Letting AI Run an Entire E-Commerce Store on Autopilot

The e-commerce playbook of the 2010s – pick a Shopify theme, hire an agency, tweak funnels, and run paid ads – is looking increasingly expensive and slow against a new generation of AI-native competitors. Amboras is positioning itself at the front of that shift. The Y Combinator Spring 2026 startup is building an autonomous AI ecommerce platform that builds, optimizes, and operates an online store with minimal human involvement.

Amboras’s promise is comprehensive. The platform stands up a store, continuously improves it, reads analytics, and learns what converts, all on its own. The company’s longer-term framing is equally ambitious: “Today it sells to humans. Tomorrow it sells to AI agents.” That second half of the pitch reflects a growing expectation among e-commerce operators that the buyer on the other side of a transaction will increasingly be an AI agent acting on behalf of a consumer – and that stores optimized for human shoppers will need to be re-tooled for machine ones.

Amboras was founded in 2025 by brothers Amin Mokadem and Imad Mokadem. Both have real e-commerce operating experience, describing themselves as former six-figure MRR e-commerce founders – which in this category is worth more than most advisory board rosters. Amin studied Computer Science at ETH Zurich, and Imad is an ETH-trained mechanical engineer. That combination of technical depth and hands-on merchant experience gives the founders an unusual read on what the AI-era storefront actually needs to do, as opposed to what a purely technical team would build.

The company is based in San Francisco and operates with a focused team of four. Its tag profile – Generative AI, B2B, and E-commerce – reflects a product that is genuinely horizontal in ambition but is being sold to businesses with a single obvious buyer: the person running or launching an online store who does not want to assemble an agency, a CRO consultant, a copywriter, and a designer to get to parity.

The market backdrop favors the pitch. E-commerce merchants have been squeezed by rising CAC, privacy-driven attribution changes, and the proliferation of tools that each solve a narrow slice of the problem. Consolidation into a single AI ecommerce platform that owns the full loop – build, optimize, analyze, adjust – is the logical next evolution. Early indications are that conversion optimization, copy iteration, and merchandising decisions are especially amenable to autonomous AI systems, because they involve tight feedback loops, clear success metrics, and structured data.

The “selling to AI agents” thesis is more speculative but increasingly credible. As consumer-facing agents become capable of comparison shopping, negotiation, and purchase execution, merchants will need stores whose APIs, content, and pricing logic are legible to machines. Amboras is wagering that a platform built from the ground up for both human and AI buyers will be better positioned than one retrofitted from a traditional storefront.

For operators tired of stitching together a dozen tools to run a single store, Amboras’s autonomous approach offers a compelling alternative. If the technology lives up to its pitch, the classic e-commerce stack may be headed for a significant compression.