Hiring a Startup Engineering Team

So, you've got your company kick-started! Perhaps you have a prototype, or even a working alpha. You might have a couple of developers already, or maybe you outsourced the initial project. You might have a seasoned technical leader, or nobody technical yet. In any case, you've got some cash in the bank, a product roadmap (you do have that, right?), and high expectations. It's time to grow, and hire a team!

Whether you're looking for one or twenty, Java or Ruby on Rails, Linux or Windows, who knows, maybe Smalltalk on a mainframe, there are some guidelines which I've collected over the years that I thought I'd share here. Hopefully others will comment and share their own ideas, or have differing viewpoints on some of mine. This is the first time I've tried writing them out like this, so we'll see how well they communicate. This will likely be a work in progress.

1. Determine short-term vs long-term needs

Hiring is a long-term exercise. Don't hire a bunch of C#/.Net developers just because that was the platform chosen by the outfit that you hired to develop the prototype. Decide what your long-term needs are, and hire for those. If your short-term needs are different, figure out why that is, and use contractors to fill the gap. Yes, smart people can learn new technologies, but startups generally have compressed timeframes and you'll need experts on hand ASAP.

2. Hire people, not skills

What I mean by this is that you're not hiring a body that happens to be able to code in RoR; you're hiring a combination of three elements: a personality, a body of experience, and a potential. You need to make sure that all three of these things fit the company you're building. This is especially important at the beginning of the process, because the first few people you hire, in addition to the founding team, will establish both precedents and the company's personality. There have been numerous occasions in the past when I've interviewed someone and known that their skills would be very valuable but they just wouldn't fit in. And while you can train someone to fill some gaps in technical knowledge, never ever believe you can change someone's personality. Just don't go there.

The body of experience defines what the person can bring to the company from their past. They may have built something similar; they may have tackled some of the challenges you face (scaling, performance); they may have worked with a similar platform or technology; and so on. This one is pretty obvious, and is what most interviews focus on.

However, I would argue that too many interviews focus too strongly on the body of experience, and not enough on the final element: potential. This is where the person can go, given the right incentive, opportunity, environment, support, and mentoring. This is also the single hardest thing to determine in an interview, unfortunately. It's naturally straight-forward to ask someone questions of a technical nature and decide if they know what they're talking about. It's much harder to ask questions and get a feel for how "smart" someone is, how curious they are, and what their potential is. It's a combination of learning why they do what they do, figuring out what gets them excited, and deciding whether that's what you can offer. Sometimes junior developers will simply need (and rightly expect) more mentoring than your current team can support. Sometimes a senior developer feels ready to move into a management or architect position, but you know that you won't have that position for the foreseeable future. Be honest with them and yourself, and you'll find the person whose desires and expectations match yours.

3. Be decisive, both ways

Be prepared to act quickly, and be decisive, both in positive and negative directions -- that is, when hiring and, alas, when firing. If you have someone come in for an interview and after they've talked to a few people everyone's nodding and saying Yes, we want this person, act then and there. When a candidate is coming in, decide beforehand what salary range you can manage, what stock options you're prepared to offer, and exactly what position you have in mind for them. At the end of the interviewing, if you like them, make the offer. Don't expect them to accept it then and there -- though that's great! -- but don't let them leave without knowing that you want them, and here's what you can offer. If you wait, they can't be sure how it went, and they may get a call with a serious offer five minutes after leaving your office.

On the flip side, if someone comes on board and you realize within a week that they're not working out, come to a decision quickly as to whether the situation can be fixed or not. Are they simply wrong for the job or the company? Did you each make a mistake in thinking there was a fit? Or are they just not getting the help they need? Figure it out, and fast. If they're just not right, don't drag it out. It might feel like it's better, but really it's not. Clear the potential disruption from your office, and let the person get on with their life and find a better position. There's never a good time to let someone go, but letting a situation drag on is never better. I've made that mistake before, and it's an easy one to keep making unless you actually enjoy firing people (which is a different problem!).

4. Try the known, settle for the unknown

When hiring, your biggest initial assets are yourself and the people already on your team. Use your network first, and encourage referrals. Offer a referral bonus to your team if someone they refer is hired (but pay after they've been with your company for at least 3-6 months). Offer a referral bonus to anyone who knows you're hiring, not just your own employees. Think of it this way: the typical recruiter will get 20% of a new hire's base salary, and you're looking at generally expensive people when hiring engineers. If you offer $2500 for referrals you're getting off very cheaply. I know of startups offering $5000 to strangers who refer new hires. It's a very competitive market out there.

Naturally, recruiters can make life easier -- if you find the right one(s). If you have the time, make a pass yourself first through the usual places. Post to craigslist and search the resumes there (but beware of the body shop phishers who post fake resumes in order to get your contact info; you'll never get rid of them). Get a Dice account and do some searching. Try posting to LinkedIn, TechCrunch, and other spots if appropriate, but keep in mind that those limit the field more than the job sites that everyone searches.

Recruiters are, to some extent, a last resort. But unless you're extraordinarily lucky, you'll be working with them. Don't pick one at random from the web -- if you haven't worked with any in the past, ask around for recommendations. There's an extraordinary number of poor recruiters who just do what you could do yourself -- search Dice and craigslist and send you the resumes. Interview the recruiter and find out how much prescreening of candidates they do and of what type. Is the recruiter technical enough to get past resume buzzwords and determine if a candidate really knows what they're claiming? At the same time, be prepared to give enough time to help the recruiter help you. Don't just read a resume for 10 seconds and delete the email. Respond with specific notes about what you like and what you don't like, because that feedback will help the recruiter understand what you're really looking for and what your priorities are. You'll start with a job description, but there are many other criteria that are hard to put in writing, from skills to personality traits.

5. Sell your company

No, I don't mean literally. What I mean is that you're not necessarily bargaining from a position of strength when you're interviewing someone. Always remember that good candidates are also interviewing you and your company. Think about how you're presenting things, from first impressions to last.

Are you a brand-new company in stealth mode? What does your web site look like? Okay, you can't say much because you're in stealth mode, but where does that leave someone who gets an email from you, when they look at your site? You'd better have something, or else you'd better make up for it with a really good introductory email. When you find a resume and contact the candidate, don't just write "We're looking for engineers for our stealth startup!" -- give them some info, hint at the niche, describe exciting technology...something to pique their interest.

When you get someone in for an interview, show your best side. If you're a startup crammed into close quarters, encourage the team to show the startup spirit. Introduce everyone, even if only a couple of people will be interviewing the candidate. Demonstrate that they'll be part of a good team if they join. Be nice, offer them water/soda/juice/coffee, make them comfortable.

Be organized -- if several people will be doing technical grilling, make sure they're not asking the same questions. Cover a good range of topics. That will not only help you get a better feel for the candidate, but it will also show the candidate that you're being thorough and organized.

Read the resume beforehand. It will be immediately clear to an interview candidate whether the person conducting the interview has taken the time to review their experience and formulate questions. Try at the very least to go through and use a highlighter to mark important items, skills, technologies. It shows interest, care, and thoroughness.

Remember: it's an extremely competitive market out there, and any good candidates -- the ones you want on your team, that is -- will have a lot to choose from. Make them want to join you and your company, and not just because you have a good idea that might make their options worth something. Make them want to join because they like you and your team and because they'll be challenged and have fun working there.