Lyft

Magic AI

Lyft

Lyft takes a commission on every ride — typically 20-25% of the fare. The driver gets the rest plus tips.

Revenue also comes from service fees charged to riders, subscription products (Lyft Pink at $9.99/month for discounted rides), and bike and scooter rentals in select cities.

The economics are straightforward but brutal. Each ride has a driver who needs to be paid enough to show up, a rider who needs a low enough price to choose Lyft over alternatives, and Lyft's cut has to cover platform costs, insurance, customer support, and hopefully generate profit.

The margins are thin — gross margins hover around 45%, and after operating costs, the company has been unprofitable for most of its existence.

Advertising is an emerging revenue stream. Lyft Media places ads on in-car tablets, the Lyft app, and bike-share stations.

It's small but growing and high-margin compared to the ride business.

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.

Lyft

Logan Green was obsessed with transportation. Growing up in Los Angeles — the car capital of America — he spent his college years studying why American cities were so car-dependent and how ride-sharing could fix it.

In 2007, at age 23, he started Zimride (named after Zimbabwe, where he'd seen communal minibus sharing), a long-distance carpooling platform for college campuses.

John Zimmer was a hospitality management student at Cornell who joined Zimride early on. The two realized that while Zimride worked for planned trips, there was no good solution for on-demand rides within a city.

Uber had launched UberCab in 2010 as a black car service, but it was expensive — a luxury product.

In 2012, Green and Zimmer pivoted Zimride into Lyft, launching a peer-to-peer ride-sharing service in San Francisco. The differentiator was branding: Lyft was friendly, casual, approachable.

Riders sat in the front seat. Cars had giant pink fuzzy mustaches (later replaced by a glowing dashboard amp).

Drivers fist-bumped passengers. It felt like getting a ride from a friend, not hailing a cab.

They eventually sold the original Zimride carpooling platform to Enterprise Rent-A-Car and went all in on Lyft.

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.

Lyft

Lyft's original growth strategy was being the anti-Uber. When Uber was mired in scandals — the Susan Fowler sexual harassment revelations, the "God View" privacy scandal, Travis Kalanick's combative leadership — Lyft positioned itself as the ethical alternative.

The #DeleteUber movement in 2017 sent a wave of riders to Lyft.

Market focus was another differentiator. While Uber expanded to 70+ countries, Lyft stayed focused on the US and Canada.

The theory was that winning one market deeply was better than spreading thin globally. This kept costs lower but also capped the growth ceiling.

Bike and scooter integration was the multimodal play. Lyft acquired Motivate (the largest bike-share operator in the US, running Citi Bike and others) in 2018 for $250 million, adding an entire transportation layer that Uber didn't have.

In dense urban areas, bikes often beat cars for short trips.

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.

Lyft

Uber is the problem that never goes away. Uber has 72% of the US ride-share market to Lyft's 28%.

Uber has global scale that generates massive data advantages, cross-selling opportunities (Uber Eats), and brand recognition. Every dollar Lyft spends on marketing, Uber can match and triple.

The market share gap has been stable for years, and closing it seems nearly impossible.

Profitability has been elusive. Lyft went public in March 2019 and lost money every quarter for nearly six years.

The company has cut staff aggressively — laying off 13% of employees in late 2022 and another 26% in April 2023. Only in the second quarter of 2024 did Lyft post its first profitable quarter as a public company.

Autonomous vehicles are both an opportunity and a threat. If self-driving technology works, it eliminates the biggest cost in ride-sharing: the human driver.

But Lyft sold its autonomous vehicle division (Level 5) to Toyota's Woven Planet in 2021 for $550 million. Now they partner with AV companies instead of building their own technology.

If Uber or Waymo crack autonomous rides first, Lyft could become irrelevant.

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.

Lyft

Lyft Rideshare — the core ride-hailing platform matching riders with drivers in 600+ cities across the US and Canada. Lyft Pink — a subscription program ($9.99/month) offering 5% off rides, priority airport pickups, free roadside assistance, and discounted bike/scooter rides.

Lyft Bikes & Scooters — micromobility options in select cities including the iconic Citi Bike system in New York City (operated by Lyft since 2018). Lyft Autonomous — partnerships with autonomous vehicle companies including Motional and May Mobility to offer self-driving rides in select markets.

Lyft Media — an advertising platform placing ads across Lyft's digital and physical touchpoints including in-app, in-car tablets, and bike-share stations.

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.

Lyft

Andreessen Horowitz led the Series A and was an early champion. Founders Fund invested early.

Fidelity, Alphabet (Google's parent), and Alibaba participated in later rounds — notably, Alphabet invested $1 billion in Lyft while simultaneously developing Waymo, a potential competitor. The March 2019 IPO raised $2.3 billion at a $24 billion valuation — Lyft beat Uber to the public markets by six weeks.

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.