Without a doubt, one of the hottest new startup accelerators in the tech industry right now is Andreessen Horowitz's Speedrun program. The accelerator, scheduled to launch in 2023, has an acceptance rate of less than 1%. The program said in a blog post in January that more than 19,000 startups had participated, but less than 0.4% were accepted into the latest group.
The program once focused on gaming-related startups, but has since expanded into entertainment and media and is now a “horizontal program,” Joshua Lu, general manager of the program and a partner at a16z, told TechCrunch. Founders of all types of startups are currently eligible to apply, and the program will run for approximately 12 weeks in San Francisco. Lu said the program used to be in Los Angeles, but will now focus on science fiction.
There are two cohorts a year, each accepting approximately 50-70 startups. The program invests up to $1 million in each company, but the downside is that it's a bit expensive. Typically, we pay $500,000 upfront in exchange for 10% of the startup through a SAFE note, and invest a further $500,000 if the next round is raised within 18 months, regardless of the terms agreed to by other investors. In contrast, Y Combinator typically acquires 7% of a company for $500,000.
Speedrun said it has more “equity value” because of what its program offers founders. This provides access to a16z's advisory and business network to assist with tasks such as go-to-market, brand development, media strategy and talent sourcing. Additionally, startups will receive benefits such as $5 million in credits to vendors such as AWS, OpenAI, Nvidia, and Deel.
Given the high interest and low acceptance rate, TechCrunch spoke to Lu for some tips on how startups can best stand out. The latest cohort begins in January and ends with Demo Day in April. Applications for the next cohort will begin in April, but off-season applications will be considered throughout the year, Lu said.
Focus on the founding team
Speedrun focuses on early stage startups. For this reason, they are seriously considering who is on the founding team and whether their skills complement each other, Lu said.
“That doesn't mean it has to be technical, it doesn't have to be commercial, it doesn't have to be marketing,” Lu says. That means, “We don't want to see obvious holes in competency or interest. We want founding teams to be aware of this and make it part of their hiring plans.”
tech crunch event
Boston, Massachusetts | June 23, 2026
I also like to see teams that I've worked with before or have a common history with.
“There's a lot that a founding team has to navigate on the startup journey, but having some pattern recognition, being able to collaborate with each other, and knowing how to come out on the other side of disagreements and disagreements are all things that people on a founding team with a common history have ease with on average,” he continued.
Even though AI has lowered the barrier to building software, Lu says having technical knowledge is still very helpful for founding teams. At the same time, Lu said the Speedrun team likes to identify when startups have already validated their product or gained some traction in the market, since AI has made it much faster to build and test hypotheses and get products out into the world.
“Speedrun as a program is very good at helping teams pour gasoline on very small sparks and fires,” he said. “We’re looking for a team that’s trying to build and try to show us there’s a little spark that can fan the fire.”
Limiting market “theory”
Lu said one common mistake founders often make during the application process is spending too much energy talking about market theory and why a clear problem exists and why the solution is the right one. “That all may be true,” he said.
At the same time, he added, even the largest and most successful technology companies have faced unforeseen obstacles in their early years, sometimes even causing them to completely pivot. What a company initially thinks of building is not necessarily what will ultimately be successful.
He continued, “What we really want to ask is why this founding team is really close. Why are they a great founding team and the best founding team to solve this particular problem?” In addition, it also validates the idea itself.
It's good to use AI in apps, but…
Lu said the program encourages all founders to use AI to “clean up” their applications. He said given the increasing sophistication of AI tools, there is no longer an excuse for grammatical errors or spelling mistakes. He also said that AI can help founders organize their thoughts and make them clearer, more concise, and more consistent.
But that could backfire if AI does all the explaining for your startup. If the founder makes it to the next round, it will be a live video call interview. “At that point, their live narration skills will be put to the test,” he said. Therefore, founders should be prepared to talk convincingly about their startup without the help of AI.
Only about 10% of founders make it to the video call stage. Typically, there are two to three investors on the review committee at a time.
After the live interview, the team typically conducts a few more screening calls with the founders, after which final decisions regarding the cohort are made.
be greedy for networking
Of course, there are other accelerator programs to choose from for startups. According to Lu, Speedrun itself was inspired by some of these other programs.
Still, he said the accelerator prides itself on providing founders with access to a large, professional operations team. In fact, he said, the best teams who get the most out of the program are those who are “the most hungry for exposure to the great people and programs” Speedrun has to offer.
Mr. Lu listed several points. a16z has about 600 employees, with 10% of its staff in the investment team, he said. All others are operators supporting companies with which the company is affiliated. As a result, Speedrun founders will have access to experts who can assist with marketing, banking, finance, administration, and many other functions. Therefore, it helps to know who startups want to connect with and why.
“We tell the founders who join the program that you get what you get out of Speedrun, you get what you put into it,” he said. “We think we are a very smart choice for founders who want to tap into the world's experts in a variety of fields in the early stages of their startup.”
Advice from the founders on the program
A recent colleague, founder Mohamed Mohamed, just announced $5 million in funding for Smart Bricks, a proptech startup led by A16Z Speedrun. He said he was drawn to the program because it stood out as one of the few “explicitly designed for co-founders working on frontier AI applications” and because he wanted a program that would allow him to “stress-test their ambitious technical visions.”
Mohamed said he treated the application more like an internal strategy memo than a sales pitch. “Rather than polishing up buzzwords, we focused on articulating the real problem, why it's structurally difficult, and why our team is unusually well-positioned to solve it,” he said. “We were clear about what worked, what didn't work, and where we needed help. I think being honest and clearly articulating why this issue was important helped the company in its application process.”
He called the entire process “rigorous but refreshingly thoughtful,” saying it was designed to understand not just what they've built so far, but how the founders are thinking. “The conversation went deep into product architecture, data strategy, and long-term ambitions. It felt more like a partner-level discussion than a typical accelerator interview, which was a strong signal for us,” he said.
His overall advice is to “be intellectually honest and accurate.” For example, he said in his application that he avoided “over-optimizing” to promote the company. “If you're vague, derivative, or overly defensive about your ideas, it will show quickly. Don't try to make yourself seem bigger than you are. Being clear about where you really are is much more persuasive than an exaggerated story,” he said.
Ultimately, he said, “Speedrun isn't looking for perfect companies. We're looking for founders who can reason clearly about complex problems and build businesses with conviction.” “Be clear about the difficult parts of what you're doing and why it's worth working on. Depth is honed every time.”