In sports, no matter how elite the athlete is, they always have a coach. Even absolute world champions have someone who will critique their form, strategize with them about the next event, encourage and push them when they are tired. Yet, in software engineering, I often see people don’t get any help and just try to power through their own career. Imagine, that you can ‘test run’ your career decision, run A/B test on it, and then make the best decision to be deployed to production. This is essentially what mentorship is about.

Q: Why would you want to have a mentor?

There are always people who have been there before. There is someone who went through the same promotion, joined the company you are considering joining, works in the domain you are interested in, built a startup, or is simply in a similar situation. You want to talk to them and learn from them. You will benefit from leveraging their experience. Instead of doing your own O(n) you can learn about optimized O(1) approaches right away.

Q: Shouldn’t you make your own mistakes?

Yes, absolutely. Make them and quickly recover, at the same time there is only so much of this life. You can live a lifetime and not make enough mistakes, and some of them are painful, especially if they are ‘one way door decisions’. So I say: make mistakes naturally while learning how to avoid them and knowing about the mistakes and lessons of others. Other people’s experience is so much cheaper to learn from than making costly mistakes on your own. I think the room for making mistakes is infinite, so, maybe, make them when it is more of ‘two way door decisions’ but get as much advice as possible for irreversible decisions.

Q: How do you get access to mentors that seem inaccessible?

Let’s say you are interested in some particular tool or technology and you know of a person who is voice in that field, say for AI that would be Andrew Ng or Andrej Karpathy. You probably won’t get 1:1 with Andrew Ng, but what you can do is find people who report to them or work with them, who appear on the same white papers as them and get in touch with those people. You can also be very specific when you reach out asking about a very specific part of their work and offering something in return. And I am not talking nonsense – I personally had small exchanges with major bloggers (back then) and with authors of some tech tools, books. What I’m saying is that people are more accessible and open than you might think. This requires high effort, but the ROI on a single response can be career-defining.

Q: Why would anyone be interested in helping you?

Although it might seem you are asking for someone’s time like if it was their charity to you, but that is not the case. There are multiple reasons why they will be interested in helping.

1. Teaching you forces them to sharpen their own thinking. This is the top of the learning pyramid. 
2. Networking. More senior engineers are looking for high-potential engineers and mentorship is often a way to build the bench and network across. It’s a win-win.
3. Sometimes it just gives people satisfaction to know that someone acted on your advice and was able to make progress.
4. A way of staying in touch with what is happening on the ground, technology stacks, career challenges, etc.

Q: How to ‘match’ your mentor?

One of the mentors I had once said that finding a good mentor is a bit like dating. Not everybody ‘clicks’ even if they seem to be the right fit on paper. In my opinion you need to have a session or two with them to understand. If you are in a big company there are official channels to establish mentorship and I encourage you to try it out. If you are looking for a mentor outside, it might be a bit weird to ask someone ‘Will you be my mentor?’ (sounds almost like a marriage proposal on a first date), but instead you probably want to ask for specific advice on some topic, share your own thoughts on the topic of their interest and build the relationship naturally. Oftentimes for mentorship to be effective you don’t have an official ‘mentorship’ label, and it can just be occasional lunch.

Q: How do you spot a good mentor?

In my opinion, a good mentor will be asking you questions that make you think a lot. They would challenge your thought process. Give new perspectives. Good mentors give you better questions than the ones you come up with.

Q: How do you prepare for the first session?

Treat the first session like a system design interview. Have requirements and constraints in mind. Explain the goals of ‘design’, give some timeframes. For example: “I want to learn how to deal with ambiguity, and this is the situation I’m in …, my goal is to close this ‘gap’ for my next promo by x”.

Q: How do you get the most out of mentorship?

Follow up! Yes – that’s the best advice I have. After each meeting with your mentor, make sure to follow up on the action items. It does multiple things: you are not wasting their or your own time; you are showing respect to your mentor; you are building discipline in yourself. Going to your mentor with the generic “How do I grow?” The question is a bit discouraging. Your mentor will have to put a smart face on and give you a generic answer. Who needs that? Go with very specific questions, create action items, report back on them, put work in. Close the loop!

Q: What’s my model for working with mentors?

I like being accountable and keeping people accountable. As such for my mentorships I keep a list of Action Items and generally strictly follow up on them. This isn’t the only valid approach. Some people find value in more casual and relaxed ‘coffee chat’ conversations.

Q: Do you need a ‘board of directors’ or just one single mentor?

I think getting more perspectives is always healthy. You need one person who is technically better than you, one who is politically savvier, and one who understands your personal values. There is, obviously, a limit to how many mentorships you can handle. Personally I have 1 or 2 mentors at any given time and besides them I have some network connections that work a bit like mentorships where I learn from them, even though there is no label ‘mentor’.

Q: When and how to end the mentorship?

Sometimes you outgrow a mentorship relationship. Let’s say you reached the goal you wanted to achieve. If you have mentorship within the bounds of a company, this is super easy to do, just tie it to the performance review cycle, thank them, leave them good feedback they can use in their perf review and move on. It is fair, a good mentor wants you to outgrow them, it’s awin-win.

If the mentorship is outside of a company, then it is more likely to have natural evolution. Never ‘ghost’ your mentor. Instead give it a logical conclusion, thank them, mention switching to more of ‘ad-hoc’ sessions instead.

Conclusion

You cannot really A/B test your entire life and career. If there are ways to move faster and learn from others then I would say mentorship is definitely one of such ways and you should leverage it. Whether you are looking for a mentor or are ready to be one, remember: we grow faster when we grow together.

P.S. This post is a ‘Thank You!’ to my current and past mentors as well as my self-reminder to leverage this even more in my life for other areas, such as my sports, personal growth, building ideas, etc.