Instacart

Magic AI

Instacart

Instacart operates as a marketplace connecting consumers with personal shoppers and grocery retailers. Revenue comes from multiple streams: delivery fees and service fees charged to consumers (typically $3.99+ per delivery), tips to shoppers (passed through, not revenue), retailer partnerships (grocers pay Instacart for access to the platform and fulfillment services), and advertising.

Advertising has become the crown jewel. Instacart Ads lets consumer packaged goods (CPG) brands like Coca-Cola, Procter & Gamble, and Nestlé pay for sponsored product placements within the Instacart shopping experience.

When someone searches for "chips," Doritos can pay to appear first. This is incredibly valuable because it's advertising at the exact moment of purchase intent.

Ad revenue was $871 million in 2023 and crossed $950 million in 2024.

The retailer partnership model is key. Unlike DoorDash or Uber Eats (which listed restaurants without permission early on), Instacart works with grocers as partners.

Over 1,500 retail banners including Costco, Kroger, Albertsons, and Publix have formal partnerships. Instacart provides the technology and shoppers.

Grocers provide inventory and stores.

Magic AI

Magic is building an AI that works like a software engineer, not a chatbot. The goal is a model you can hand a whole code base and a task, and it writes, reviews, and debugs like a teammate.

Their edge is memory. In 2024 they showed off LTM-2-mini, a model with a 100 million token context window.

To put that in perspective, that is about 10 million lines of code the model can look at in one shot, by far the biggest of any model at the time. The plan is to sell this as an AI coworker to companies that write a lot of software.

As of the 2024 rounds, the product was still not for sale.

Instacart

Apoorva Mehta was a 26-year-old Amazon engineer in Seattle who quit his job in 2012 to start a company. The only problem: he had no idea what to build.

Over the next year, he started and abandoned roughly 20 different projects. A social network for lawyers.

A way to track restaurant wait times. Nothing stuck.

Then one day he was too lazy to go grocery shopping. He looked for a service that would shop for him and deliver everything to his door.

Nothing good existed. The existing options were grocery store delivery services that only worked during specific windows, had limited selection, and required ordering days in advance.

Mehta wanted to order groceries the way he ordered everything else online — immediately, from whatever store he wanted.

He built a prototype in 2012 and applied to Y Combinator. The demo was rough — he ordered a six-pack of beer through the app and had it delivered to a YC partner's house during the application process.

It worked. He got in.

Instacart launched in the San Francisco Bay Area in 2013 with a simple promise: order from your favorite local grocery store and have someone shop for you and deliver within an hour.

Magic AI

Eric Steinberger and Sebastian De Ro started Magic in 2022. Steinberger is an AI researcher who became convinced that the missing piece for a true AI software engineer was memory.

Chatbots forget what you told them a few thousand words ago. He wanted a model that could read your entire code base and remember all of it.

They raised a small seed in 2022, then a $23 million Series A in early 2023 led by Alphabet's growth fund CapitalG. From day one the pitch was not a chatbot.

It was an AI colleague that could handle real engineering work.

Instacart

Instacart grew by solving a problem one city at a time. They launched in San Francisco, proved the model, then expanded to other major metros.

Each new market required recruiting shoppers, signing up retailers, and building enough consumer density to make the economics work.

The COVID-19 pandemic was the inflection point. Grocery delivery went from luxury to necessity overnight.

In March 2020 Instacart announced it would add 300,000 shoppers, then raised that by another 250,000 weeks later. Order volume increased 500%.

Years of planned growth happened in weeks. The pandemic proved that grocery delivery wasn't a niche — it was the future of how a significant chunk of the population would shop.

The enterprise play is the long-term moat. By providing white-label technology to grocers, Instacart becomes embedded in their operations.

Even if a grocery chain wanted to build its own delivery service, they'd need years and hundreds of millions to replicate what Instacart provides. The more deeply integrated Instacart becomes in grocery operations, the harder it is to rip out.

Magic AI

Magic's strategy is research first, product later, which is both unusual and expensive. Instead of shipping an app and iterating in public, they raised huge sums to chase a technical breakthrough in model memory.

The headline move was the 100 million token context claim, which got the whole AI world talking. They also cut a deal with Google Cloud in 2024 to build two supercomputers, called Magic-G4 and Magic-G5, using NVIDIA chips, to train bigger models.

Star backers like Eric Schmidt served as validation. The bet is that if the tech is far enough ahead, the product and the customers follow.

Instacart

The post-COVID hangover was brutal. After pandemic demand normalized, growth slowed dramatically.

The company's valuation dropped from a peak of $39 billion in early 2021 to about $10 billion at IPO in September 2023. Investors who bought at the peak saw a 75% paper loss.

The narrative shifted from "essential infrastructure" to "nice-to-have luxury."

Unit economics are perpetually tight. Paying a person to walk through a grocery store, pick items, bag them, and drive them to someone's house is expensive.

Unlike meal delivery (one restaurant, one bag), grocery delivery involves dozens of items per order, refrigeration requirements, and substitution decisions. Every order that requires a shopper to call the customer about an out-of-stock item eats into efficiency.

Amazon is the existential threat. Amazon Fresh, Whole Foods delivery, and Amazon's own logistics network represent a competitor with nearly unlimited resources and a Prime membership base of 200+ million.

Amazon has been willing to lose billions on grocery delivery to build market share. Instacart's advantage is retailer partnerships — Kroger and Publix use Instacart specifically because they don't want to help Amazon dominate grocery.

Magic AI

Magic has raised close to half a billion dollars and, years in, still has no product the public can buy. That is the whole problem in one sentence.

A giant context window is a great demo, but rivals like Cursor, GitHub Copilot, and Cognition's Devin already have real users and revenue. Magic has to prove that more memory actually makes a smarter engineer, not just a bigger number on a slide.

And it has to do that while burning enormous sums on compute against competitors who are already shipping. The clock and the cash both keep running.

Instacart

Instacart Marketplace — the core platform where consumers order groceries from 80,000+ stores for delivery or pickup, with personal shoppers fulfilling orders. Instacart+ — subscription service ($9.99/month) offering free delivery on orders over $35, reduced service fees, and credit back on pickup orders.

Instacart Ads — a retail media platform letting CPG brands run sponsored product listings, display ads, and coupons within the shopping experience. Instacart Platform (Enterprise) — white-label e-commerce technology that lets grocers build their own online ordering and fulfillment powered by Instacart's infrastructure.

Caper Cart — AI-powered smart shopping carts (from the 2021 Caper AI acquisition) with built-in screens, barcode scanners, and payment that let shoppers skip the checkout line.

Magic AI

The centerpiece is Magic's LTM, or Long-Term Memory, model family. In 2024 they unveiled LTM-2-mini, a model with a 100 million token context window, pitched as the key to an AI that can understand an entire code base at once.

The broader product is an AI software engineer, or coworker, that can write and change code across a whole project. The company also announced Magic-G4 and Magic-G5, two supercomputers built with Google Cloud and NVIDIA to train these models.

Much of it is still research rather than a shipping product.

Instacart

Sequoia Capital was an early and consistent backer. Andreessen Horowitz invested in growth rounds.

D1 Capital Partners led the 2021 round that valued Instacart at $39 billion. Existing investors including Valiant Capital, T.

Rowe Price, Fidelity, and Tiger Global participated across rounds. Y Combinator was the starting point (Summer 2012 batch).

The September 2023 IPO on NASDAQ priced at $30 per share, valuing the company at approximately $10 billion.

Magic AI

Magic's backers are a murderers' row of AI insiders. It started with a $5 million seed round in 2022 and a $23 million Series A in early 2023 led by Alphabet's CapitalG.

Then in February 2024 came a $117 million round led by Nat Friedman and Daniel Gross, two of the most connected investors in AI. The blowout was August 2024.

Magic raised $320 million with checks from former Google CEO Eric Schmidt, CapitalG again, Atlassian, Jane Street, Sequoia, and Elad Gil. That pushed total funding to around $465 million, all for a company still building toward its first sellable product.