I wrote this post for myself at the end of 2020. A few friends read it and told me they found it useful, so I decided to publish it.
I've broken into three new fields in my life as an outsider: physics, software engineering, and startups. I'm currently trying to break into a fourth. I'll let you know how that turns out. 
I've noticed a common pattern when I've tried to do this. The pattern is that, at first, I'm always weighed down by the crushing burden of my own incompetence. I call this phase the Muddle, and I haven't figured out how to avoid it.
The Muddle is when everything you do is wrong, upside down, and backwards. The Muddle is when you can't tell which way is up, or even who to ask for directions. The Muddle feels like you're climbing a blank wall with no handholds. Whatever the hell "traction" is, the Muddle is the exact opposite.
On the far side of the Muddle you'll find competence, confidence, and success. But there's no guarantee you'll ever reach them. So the Muddle is fear of failure, too.
I've found that the people who can get through the Muddle have two qualities. First, they're irrationally optimistic that they can break in quickly, despite overwhelming evidence to the contrary. And second, when their first attempt at breaking in ends in a humiliating catastrophe, they quickly integrate the lessons from that failure and cheerfully apply them to their next irrational attempt. After they've gone through enough cycles of overconfidence and disaster, the world runs out of lessons to teach them. In other words: they've broken in.
It's an odd mix, isn't it? Fast learning at the object level, yet exasperating subbornness at the meta level. You can see why people would want to go through this process, over the course of their lives, somewhere between once and zero times.
How to choose
It takes a lot of learning cycles to break in, and that means it also takes a long time. So it's worth considering how you can avoid wasting your time on a field that might turn out to not be very good. There are a few questions you can ask to tell beforehand if a field is worth the time.
First, and most important: is the field welcoming? If you notice the participants act in territorial ways, that probably means the field has peaked, and is in a negative-sum decline. Avoid it. A field that's worth breaking into will be trending up, not down. In a good field, people can sense the opportunity, and they tend to act in positive-sum ways as a result. So if you notice there's a field where people are unusually welcoming, that's a sign that it might be worth looking into. Happily, you can follow your instincts here: unhelpful behavior should rightly warn you away from any field that tolerates it.
Second: do there seem to be a lot of smart people in the field? By this, I don't mean that everyone in the field is smart. I just mean that the density of smart people seems noticeably above the average. Smart people tend to sort themselves into the fields where the good opportunities are, so their presence in a field is a sign that it's promising.
And third: do you feel so strongly drawn to a field that you're inclined to ignore all warning signs about it? If the answer is yes, then that's the easiest decision of all. You should just go for it, regardless of what anyone else says.
So a good field is one that's growing fast, and that's full of smart and helpful people. This has a lot of benefits. Apart from being nice to work with, smart and helpful people are the best kind to learn from. And a field that's growing is one that lets you accumulate experience surprisingly fast, at least on a relative scale. If the number of people who work in nanotechnology is tripling every year, and you join the field today, a year from now you'll be more experienced than 2/3 of all nanotechnologists. And two years from now, you'll be more experienced than 8/9 of them.
Once you've decided you want to break into a field, you'll need to do two things. The first is emailing strangers. And the second is impressing those strangers with your phenomenal speed.
People are flattered when you try to break into their field, so emailing strangers isn't quite the uphill battle you might imagine. In a well-chosen field, you might be surprised at how willing people are to help complete strangers. And if you happen to have broken into other fields in the past, you might be able to leverage that past accomplishment to break into this one.
When building new connections, your iteration speed is your most important asset. Ideally you should be able to follow up with someone every few days with a significant increment of progress — whether it's an insight, a piece of work, a question that indicates substantial progression from your previous position, or a report of something you've published. It's very similar to a startup: you're trying to create FOMO with your pace of progress, and surprise your contacts with how much your understanding has deepened, how much you've gotten done, or how many other people they know are helping you.
You'll sometimes burn your rep with your first contact in a field to build a better rep with your second contact. Your first contact might get exhausted with how slowly you learn, because they've forgotten how much time it took them to get to where they are. This happens to a lot of people, and it's okay. You can recover from this by leveraging everything you learned from your first contact to impress your second contact. The impression you're trying to create is "Whoa, this person might actually become my colleague one day. I'd better hedge my bets and be nice to them now." If enough people believe this, and you leverage their help with enough focused intensity, that belief will magically come true.
I've found goal coherence time is an important concept to remember when you're breaking into a new field. At first, you'll only have stable goals on a timescale of a few days. In the beginning, you're pretty likely to discover something today that will totally change your plans for tomorrow. For example, you might discover that the person you were going to work really hard to get in touch with isn't such an important node after all, and you should be reaching out to someone else instead. So at first, it makes no sense to plan more than a day or so in advance.
Once you're a month or two into the process, that will change. You're less and less likely to have your worldview upended on any given day. As a result, your goal coherence time increases. Within a month or two, it might make sense to set weekly goals for yourself, in the expectations that those goals will mostly remain stable. That's progress.
As you go deeper into a field, your goals should become more stable, and you should start thinking on longer timescales. After 3-6 months, you should be setting goals on daily, weekly, and monthly cadences. As you get entrenched (Series A for a startup, deep comfort level in a field for an individual), you can add quarterly and annual cadences too. Your longest-cadence check-in should be reserved for fine tuning your long-term vision of what you want.
Keeping your spirits up
You won't get much positive feedback early on, so don't expect it. This is by far the hardest part of breaking in, and it's insidious: over time, the lack of positive feedback starts to weigh on you like an invisible anchor that slows down everything you do. You can succumb to this without even being aware of it. In fact, the only time I realize how painful the lack of feedback is, is in the rare case someone gives me a compliment: I feel light as a feather for days afterwards. When trying to break in, I become a compliment camel, treasuring every speck of praise I can find in a desert of indifference.
The way around this is to constantly ship visible things. If you're working on something hard or deeply technical (and therefore long-term), switch every once in a while to working on something visible. You want to preserve in your mind the link between doing work, and getting rewarded. Otherwise your mind will flinch away from work, towards other activities like Netflix or social media. Then you won't get anything done — in fact, the escalation of this cycle is what we call burnout.
No matter where you are, there's always a level above your own. In the end, nobody every breaks in completely. But we all muddle through.
 Writing from 2023, I can say it's turned out well.