How to research a startup before you interview

Before you interview at a startup, spend 30 minutes checking five things: how recently and how much it raised, who the founders and investors are, whether the product has real users, what its public signals say, and whether the funding it claims actually exists. You can confirm all of it from public sources without talking to a single person. The point is not to find a perfect company. It is to walk into the interview already knowing what to ask, and to avoid spending two weeks on a process for a company that is quietly running out of money.

Most engineers research the role and skip the company, then learn the hard truths in week three of the loop or, worse, three months after signing. The research below front-loads that. Every item is something you can check yourself today, and most of it sorts the field faster than a recruiter screen will.

How much time should this take?

For a company you are mildly curious about, 15 minutes. For one you would actually leave your job for, an hour. The first pass is a filter: funding, founders, and one traction signal. If those clear, you go deeper. Doing it in that order means you stop early on the companies that were never going to make the cut, which is most of them.

1. Funding and investors: how recent, how much, from whom

Funding recency is the single most useful signal you can get, and it is the one the roles.cc board sorts by. A company that closed a round in the last few months has runway, an approved hiring plan, and someone with more information who just priced the business. A company that raised 30 months ago and has gone quiet is a different bet entirely. We wrote about why this signal beats almost everything else in why funding recency is the best hiring signal.

Three numbers tell you most of the story:

• Last round and date. A $12,000,000 Series A closed four months ago means roughly 18 to 24 months of runway and active hiring. The same round closed 28 months ago means they are probably raising again or cutting. Check the date, not just the amount.
• Amount and stage together. A $3,000,000 seed and a $40,000,000 Series B are different companies with different expectations of you. Match the stage to the role you want. Our guide to startup funding rounds explained for job seekers maps each stage to what it means for your job.
• Who led it. A lead investor with a track record signals diligence already happened. You are not the first person to vet this company. Name investors are easy to look up and worth two minutes.

Where to look: Crunchbase (free tier shows the last round for most companies), the company's own announcement post, and the press coverage that ran when they raised. Our recent raises page lists companies by how recently they closed, so you can start from the raise and work backward to the role.

2. Founders and team: who is running this?

At a 15-person startup you are joining the founders as much as the company. Pull up their LinkedIn and read it like a resume you are evaluating. You are looking for three things: have they built and shipped before, do they know this specific problem, and have they kept people around.

• Prior shipping. A founder who has shipped real products, startup or not, de-risks the company more than a founder with only a deck. Repeat founders are not always better, but they make fewer first-time mistakes.
• Domain fit. A founder selling to hospitals who spent eight years inside one knows things a smart generalist does not. Domain fit shows up in how specific they are about the problem.
• Retention. Look at who joined early and whether they are still there. A founding team that has churned three heads of engineering in two years is telling you something. So is a team that has held its first five hires for three years.

Also check the engineering headcount and how fast it is growing. A team going from 6 to 9 engineers after a raise is a normal post-funding ramp. A team that 4x'd in six months and is hiring you into the chaos is a different job. If the role is a first or early engineering hire, read what a founding engineer actually does before the call so you can tell whether the title matches the work.

3. Product and traction: do they have real users?

A raise validates that investors believed the pitch. It does not validate that customers are paying. Those are different facts, and the second one matters more for whether your equity is ever worth anything. Spend ten minutes finding evidence of actual usage.

• Use the product if you can. Sign up for the free tier. A polished onboarding and a product that works tells you more than any careers page. A signup flow that breaks tells you something too.
• Look for named customers. Case studies, logos on the homepage, and customer quotes are claims you can sometimes verify by searching the customer's name plus the startup. Logos with no stories behind them are weaker than they look.
• Check review sites and communities. G2, app stores, Reddit, and Hacker News threads surface what users actually say. One angry thread is noise. A pattern is signal.
• Read the changelog or release notes. A product shipping visible updates every couple of weeks is alive. One that last shipped eight months ago, on a fresh raise, is a flag worth raising in the interview.

4. Public signals: what the internet says when no one is selling you

Press releases are written by the company. These signals are not, which makes them more honest. None is decisive alone, but together they correct the picture a careers page paints.

Glassdoor deserves a note. Ignore the star average and read the body of the 2 and 3 star reviews, where people are specific without being unhinged. If three reviewers independently mention the same problem (the same manager, the same chaos, the same broken promise on equity), believe it. If the complaints are scattered and generic, discount them.

5. Form D filings: confirm the money is real

This is the step almost nobody takes, and it is the one that catches inflated claims. When a US company raises money from investors, it usually files a Form D with the SEC within 15 days of the first sale. The filing is public, free, and hard to fake. It lists the company, the date, and the total amount raised in that offering.

Go to SEC EDGAR full-text search (or just search 'SEC EDGAR' and use the company-name lookup) and type the legal entity name. A matching Form D with a recent date and an amount near what the company announced confirms the raise happened. A press release announcing a $20,000,000 round with no corresponding Form D, or a Form D showing a much smaller number, is a reason to ask careful questions. Funding recency is the whole thesis behind how we sort roles, and Form D is one public source we read for it. More on that in signals a startup is about to hire.

How to turn 30 minutes of research into interview questions

Research is only useful if it changes what you do. Every flag you find converts into a question you ask on the call. Found a 22-month-old round? Ask about current runway and the next raise. Saw engineering shrink? Ask why people left and who. Could not find paying customers? Ask for the revenue or usage story in plain terms. A candidate who asks specific, informed questions reads as senior, and the answers tell you more than the careers page ever will. We collected the best of these in questions to ask in a startup interview.

If you are not sure which companies are worth researching in the first place, start from the raise. Sorting by funding recency narrows the field to companies that have money and a reason to hire right now, which is most of what this research is trying to establish anyway. Our guide to how to find startups that are hiring covers where to start the search before you start the digging.