How AI is influencing early roles at startups

AI promises increased productivity at a lower cost. For startups, that’s the dream: do more with less. In addition to rethinking the tools needed to get work done, AI is also impacting how startups approach hiring.

What is the role of junior hires, if many of the tasks can be handled by AI? What is the right time to bring on a dedicated AI role, like a Chief AI Officer or AI Operations Lead? How do you build a team when AI’s capabilities are changing so rapidly?

There’s no clear answer — but there are some considerations that all founders can ponder when they think about hiring in a world where AI is increasingly relevant to everyday work.

For a long time, startup hiring followed some familiar patterns. You'd bring on experts in a few areas (a head of sales, a CTO, a CFO) and then add in some generalists to get the work done, like software engineers and customer support reps. Outside of the experts, many roles could be filled by more junior people. These employees, often eager to learn, can grow with the company and learn from those with more experience. 

But now, AI can handle a lot of the work that used to land on a junior employee's to-do list: data entry, first drafts, basic research, and routine customer support. It has many people wondering if AI is replacing entry-level jobs traditionally filled by college graduates and those earlier in their careers. 

Dan Shipper, co-founder and CEO of Every, said the following about the impact of AI on jobs on Lenny’s Podcast:

"There's this huge question about what happens when entry-level jobs are taken away by AI. My take is that [it’s] worth thinking about, and it's possible that [it] might be a problem at some point… But people are going to figure out that some 20-year-old with a ChatGPT subscription is super powerful if you just mentor them."

In Shipper's mind, the concern isn't playing out quite as expected. Rather than eliminating junior roles entirely, AI can actually amplify what junior employees are able to do — turning them into more capable contributors, faster. As he sees it, the immediate question isn't whether companies should still hire early-career talent, but how they can help junior team members — and all team members, really — work alongside AI tools to deliver more value.

Every's own structure reflects this thinking. The company is relying on AI for everything from editing its high-quality content to coding its products. The company even has a dedicated AI Editorial Operations Lead, Katie Parrott. She describes her role as focused on building AI workflows that support publishing Every’s newsletter and marketing messaging.

Parrott says that Every has a really important principle in its approach to AI: don’t repeat yourself. “If you notice you’re doing something over and over again, that’s something that’s a prime use case for automation, especially with AI.”

While AI can do a lot, it can't do everything. The full AI impact on employment remains to be seen, but strategy, ambiguity, creative leaps, and customer conversations still require human judgment.

And there’s another category that’s often overlooked: the ongoing human involvement to make AI systems work to their full potential. Alex Duffy, a contributing writer at Every, says AI is more like playing an instrument than using a tool. It takes practice, and everyone’s experience is going to be different. 

That’s where Parrott’s role comes into play at Every. Her job includes maintaining and documenting AI systems. “People think it’s a one-and-done process,” she says, “That you can build something, launch it, and that’ll be it. And it’s not.” That ongoing tweaking and testing of new AI models relies on human interaction — someone who understands the business and potential use cases. 

A combination of subject matter expertise and AI resourcefulness is the new hiring unicorn for startups. You need people who understand the work deeply and can figure out how to make AI do it well.

For startups to truly embrace AI’s potential, AI adoption can’t be something that happens on the side. It has to be built into how the team operates.

That means evaluating how every member of the team is using AI and how AI can improve their work. 

Hire a dedicated AI role within the company

While you may not need a dedicated AI role at your startup immediately, it may be a hire you want to consider sooner rather than later. “If you’re looking to operationalize AI at scale, it takes full-time resources,” says Parrott. “You need someone whose job it is to build systems for the people who are too busy doing otherwise.”

The overhead needed to manage AI systems doesn’t need to be distributed across an entire team. It should live within a centralized owner who has time budgeted in — whether it’s part of their job or their entire job.

Give people the space to experiment

While you may have a dedicated role for managing your AI systems, you’ll want to ensure that all hires are comfortable using AI. That includes budgeting time into the workweek to let people experiment.

Every lets each individual on its team choose the AI tools that work best for them, whether it's writing or coding. The company also gives the team time to experiment. “I wouldn’t have been able to make progress without the permission at Every to try things,” says Parrott. “We experiment even if we don’t know what the outcome is going to be — because we often don’t know what the outcome is going to be when we’re working with nascent technologies that are changing all the time.”

Encourage sharing, and incentivize adoption

Even with the right systems in place and encouragement to experiment, adoption doesn't happen automatically. The AI impact on jobs that are remote or in-office should be the same, because you will give people the space to share. Every uses forums for sharing, including weekly standups, Slack channels, and show-and-tell sessions. Employees discuss things they’ve tried or prompts that have gotten really good results. 

You could even create incentives — such as prizes — for people who build the coolest AI solutions. External incentives can help people get over the "I have other work to do" hurdle and can motivate people to solve real problems for the team.

While every startup’s approach to AI will be a bit different depending on your product and how your business is structured, there are a few guiding principles you can keep in mind:

1. Consider your needs when you write job descriptions

Before posting any new role, think about what tasks are actually repetitive and could be handled by AI. Then think about what requires human judgment, context, or creativity. The answer to "Do we need a junior marketing hire?" might actually be "We need a senior marketer who is comfortable with AI.” 

You’ll also want to think about the point at which you hire an AI-focused role and what that job description looks like for the role within your company. Since it’s a new role for a lot of companies, you may not have many job description examples to draw from. 

Generative AI company Writer describes a potential role as an “AI program director” who may be responsible for:

• Implementing AI in a way that serves your business goals
• Choosing where to implement AI processes
• Finding the right AI tools for your organization 
• Measuring and reporting on AI workforce impact
• Making sure your team can successfully use AI
• Ensuring your team uses AI ethically and responsibly

2. Budget for both tools and training

AI tools require an upfront investment. Not only do you need to pay for the tools themselves, but you need to budget the time necessary to build prompts, test workflows, and iterate. 

The experimenting time that Parrott described has to be allocated intentionally. It won't happen if everyone is heads-down on day-to-day tasks. “People will get frustrated in the beginning when they try something, and it doesn’t work,” Parrott points out. “You have to take the time to give feedback, show examples, and document the use cases you want to abide by. But it pays off in the long run.”

3. Hire for existing experience (plus the ability to learn)

AI’s impact on the job market creates challenges for new graduates in unexpected ways. Rather than AI directly taking over entry-level work, companies are increasingly opting to hire mid-level or senior people who can use AI to accomplish what would have previously required a small team.

Mid-level or senior-level hires who've done the work themselves can direct AI effectively. Subject matter expertise plus AI resourcefulness is the combination to look for. Employees need a certain amount of “reps” under their belts to direct AI models, identify the outcome they’re looking for, and set standards to ensure AI systems don’t go rogue.

That said, junior talent still has a place if you're willing to invest in mentorship — and there are good reasons to do it. Again, junior hires typically cost less and can grow into roles tailored to your specific needs and workflows. When experienced team members can guide them in both the craft and how to use AI effectively, you can build capable contributors faster than before. The key is having the support structure in place.

4. Plan for ongoing tweaks to AI systems and workflows

Using AI is often personal and very individualistic. Everyone develops their own ways of prompting and interacting with the tools they use. 

Because the models are always changing, your business needs to develop a process to stay on top of new releases. Remember when ChatGPT couldn’t search the internet? Then one day it could, and every use case had to adapt.

“You’re doing yourself and your business a disservice if you don’t try new things as they come out,” says Parrott. “You have to build systems that are open-ended enough that when a new opportunity, new systems, or new models come along, you’re flexible enough to adapt.”

5. Measure what matters

Quantitative metrics matter: Are people spending less time on repetitive tasks? Is output volume increasing? Are projects getting completed faster?

But qualitative signals are just as important: Are AI recommendations actually making it into the final work? Do team members find the tools genuinely helpful, or are they working around them? Is the quality of output improving?

In addition to standups or show-and-tell sessions, find ways to survey the team and measure the ROI of your investments.

AI has expanded what a small team can accomplish. Founders should think less about filling traditional roles and more about how work can get done with AI tools. Then, they need to hire people with the right skills and tenacity to put AI to work in a way that truly makes a difference in the organization.

The answer might be fewer people with more experience, augmented by AI. It might be a dedicated AI operations hire. It might be fractional consultants combined with better tooling. The hiring playbook is being rewritten in real time. The faster you figure it out, the more equipped you’ll be to keep up with advancements in AI technology.