.svg)
If you’re new here, this is the latest edition of the Build Mode newsletter, where we gather the collective wisdom of the people building with AI, designing the future of work, and leading the most important companies of the next decade. Subscribe here to get the top insights in your inbox every week.
The Big Idea: What does it mean to be a software engineer in the age of Devin?
When the startup Cognition introduced Devin, “the first AI software engineer,” last week, we felt again that particular sensation that comes with product launches in the age of AI: a mixture of excitement, awe, dread, and skepticism.
If we go by the claims in the demo, Devin has passed engineering interviews from leading AI companies, completed real jobs on Upwork, and passed 13.86% of test use cases in the set of benchmark problems, which is way above the mainstream LLMs.
Ethan Mollick tested it out and said, “If you are used to chatbots, working with Devin can feel like seeing the future.” Raphael Ouzan, CEO and co-founder of A.Team, tried it out as well and said it was “heavy but promising.”
One of our core tenets here at Build Mode is that anything that can be commoditized will be, coding including. This idea has been taken to the extreme by Jensen Huang, CEO of Nvidia, who has said that everyone in the world is now a programmer.
Some coders aren’t so sanguine. In an essay for The New Yorker, James Somers, a developer at Genius, lamented the waning days of the craft. “My wife is pregnant with our first child,” he wrote back in November. “I code professionally, but, by the time that child can type, coding as a valuable skill might have faded from the world."
Gergely Orosz isn't convinced: “Press reports say Devin is ready to do the job of developers, but even Cognition admits that the tool only solved about 1 in 7 GitHub issues unassisted in tests. That’s impressive, but there’s a very long way to go!”
Whether or basic coding will become obsolete, there’s a more interesting question here: What will the software engineer of the future look like? How will the job change now that non-technical people can build products?
To get to the bottom of this, we caught up with Maxx Metti, A.Team’s resident AI expert who got his PhD in Computational Mathematics at UCSD. He’s bullish on what this means for seasoned developers.
“If we look at no-code products, we see that they work great for PoCs and MVPs. But building scalable solutions is still not quite there. I can easily imagine in the next 2-4 years a single senior dev equipped with Gen AI will be as effective as that same senior dev with a team of 2-3 junior developers.”
Junior devs, in other words, might be in trouble. “It will be a split. The ones who keep operating as if nothing has changed will probably find themselves as replaceable. The ones who are careful to hone their craft and leverage Gen AI will succeed.”
CHART OF THE WEEK
Engagement at work among elder Millennials is in ‘dramatic decline’
Since the pandemic there has been a major reversal in engagement at work trends between age groups. Millennials and Gen Z used to be 5-6 points more engaged than Boomers and Gen Xers. Everyone has lost ground since then—except for Boomers who are more engaged than ever.
According to a recent Gallup study, elder Millennials are in dramatic decline, having dropped form 39% to 32% in the last three years.
The report found that “Millennials and Gen Z employees have seen the greatest decline in feeling cared about by someone at work, having opportunities to learn and grow, feeling connected to the mission of the organization, having progress discussions with managers, being given opportunities to develop, and feeling that their opinions count.”
Rex Woodbury drove this point home in his Digital Native Substack last week: "We’re seeing a shift to more flexible, self-driven work… Gen Z in particular wants to forgo traditional forms of work in favor of more autonomous options.”
We found similar themes in our latest research on The Great Betrayal: Respondents under 45 were much more likely to say they felt less secure committing to one employer (66%) than respondents 45 and up (50%).
You can read more about how The Great Betrayal is impacting workers in Fast Company.
AROUND THE WATERCOOLER
Where is Sora getting all that training data?
Sora is impressive—even if some of the demo videos seem to abide by a different set of physics than we do here on earth.
How did it get so good? By training on vast amounts of data. Where did all that data come from? “Was it YouTube?” A simple question posed by the Wall Street Journal’s Joanna Stern to OpenAI’s CTO, Mira Murati, who, of all people, definitely knows. Which is why she was getting dragged on X for claiming she didn’t.
In Murati’s defense, OpenAI is getting sued up the wazoo. This kind of evasiveness is standard procedure. But the question still stands.
As Dan Balsam explained when we Slacked him about this: “It was an awkwardly phrased response but my guess is that the policy is never to reveal the training data because it’s a lose-lose. It’s their main proprietary edge over other labs and they’ll get sued which will be annoying and expensive even if they ultimately win.”
The only way that training on YouTube videos would expose them to a lawsuit was if they violated YouTube’s Terms of Service. And the question of whether or not training on copyrighted material counts as copyright infringement is currently being decided by the courts.
This lends itself to a broader—as yet unanswered—question: Is your data your advantage or not?
WOMEN IN AI
3 Women on Dismantling Gender Bias in AI
In the world of AI development, there’s clear evidence of entrenched bias against women, which prevents female pioneers in the field from getting the recognition they deserve. We wanted to hear from the women building with AI and get their take on why this bias exists and how to dismantle it.
Yael Burla, Principal Product Marketer at Vimeo, on why correcting bias in AI isn’t easy
“Unless the technology is built with the guardrails and awareness of biases that exist within us as humans, women and minorities more broadly will continue to be impacted through underrepresentation and discrimination.
"Remember the classic example in 2018 when Amazon’s resume scraping tool discriminated against women? The algorithm was trained on existing top performers, favoring resumes with words that men tend to use, like “executed” and “captured,” while downgrading terms centered around women as well as graduates from all women’s colleges. Naturally, this only further worsened the problem of promoting the dominance of white males in technical and senior roles.”
Sarah Bird, Global Lead for Responsible Engineering at Microsoft, on how AI presents an opportunity to right old wrongs
"I was drawn to AI because I believe it has the potential to make the world a better place by creating solutions that are designed for individuals, rather than a one-size-fits-all approach.
"Historically, many solutions, including medicine, were not designed with women in mind. AI has the potential to change that. AI is the perfect bridge between people and technology, and it requires a deep understanding of both to create effective solutions. Innovating across these two areas is incredibly rewarding because both are interesting and important."
EVENTS
Celebrate the Women Shaping AI Next Wednesday in NYC
The Women Shaping AI Salon is happening next week!
We’ll be gathering with a highly-curated community of NYC’s top tech leaders from 5-8 pm at the A.Team Clubhouse in NYC to celebrate the women leading the AI revolution — together, we'll delve into the impact of AI on user research, strategic enterprise integration, the evolution of SaaS, and the ethical dimensions of the technology.
There are limited in-person seats available, reserve your spot now to join us.
DISCOVERY ZONE
If you find yourself turning to reddit for answers try GigaBrain. This AI tool scans through billions of reddit comments to find the no-BS, niche info you’re searching for.
MEME
.jpeg)
.jpg)
