Rust or Machine Learning: Which New Skill to Prioritize for Job Market Edge

[maara]

I'm considering learning a new skill to enhance my job market prospects. Should I invest my time in becoming proficient in Rust, known for its system-level programming capabilities, or dive into the world of Machine Learning, with its growing demand across various industries?

I'd also like to know which of these skills should be prioritized for a quicker edge in the job market. Additionally, how long does it typically take to ramp up in either of these fields? Your insights and experiences would be greatly appreciated!

[qd011]

You have an edge in the job market for things you've done before. When you jump into a new language/domain, you're at a disadvantage - it's the exact opposite of having an edge.

Doesn't mean that you shouldn't do it but the question you should be asking yourself is more like which of the two you want to do, whether it's realistic to overcome the disadvantage you're putting yourself at, and whether the switching cost is worth it.

This is assuming that you're considering a language/domain switch; if you want to keep working on whatever you're working now and think that adding Rust or ML to your CV will help you - it probably won't.

[maara → qd011]

Given my experience spread across building distributed real-time systems and data pipelines, I want to either invest getting proficiency in Rust for building tier-0 services such as database and infrastructure components, or expand my skillset in data engineering to AI/ML.

[baristaGeek]

It really comes down to personal preference rather than the trends in the job market when comparing those two in-demand skills.

Personally, I'm more drawn to "applied AI engineering" – it's like a blend between data science and product engineering, encompassing areas like vector databases, prompt engineering, agents, chains, and so on. Meanwhile, I'm also picking up modern high-performance languages like Rust and Golang on the side.

[maara → baristaGeek]

Interesting. I do read more about Applied AI engineering.

[luigi23 → baristaGeek]

Any sources?

[giardini]

One way to look at your choice is to ask which path will expose you to the most new ideas.

If you're already proficient in one or more procedural programming languages then certainly Rust training will fine-tune your skills. OTOH if you have no procedural programming skills then Rust (or C et al) would be a good place to start if you want to learn to program.

To study machine learning is a completely different endeavor that (gracefully) begins with high-school level math (algebra, matrices, linear algebra) and ends at a research frontier where, to all appearances, no one fully understands what they're doing, but are convinced it works! And since demand is endless, "there's gold in them thar hills!"

I vote for ML but take your time and don't get lost searching for gold in the hills. Maybe look for side-opportunities as did Levi Strauss:

https://www.thoughtco.com/levi-strauss-1992452

https://duckduckgo.com/?q=levi+strauss+gold+rush

[marvin-hansen]

Why not both?

Rust is a great choice for anything data science. https://deepcausality.com/blog/views-on-rust-ml/

Also, hugging faces released a new ML framework for Rust. https://medium.com/@Aaron0928/hugging-face-has-written-a-new...

There are Rust bindings for TensorFlow and PyTorch. Just google it. There is enormous value in Rust ML.

[maara → marvin-hansen]

I appreciate you for sharing your insights on Rust ML. I definitely feel that option exists too but can't have two things going at the same time.

[mortylen]

As already described, it depends mainly on what you enjoy, so that you don't get tired of learning new things. As far as the job market is concerned, knowing only one skill is not an advantage, it is best to excel in a certain combination of skills. And it seems to me that Rust and Machine Learning go together. You can learn the basics of Rust quickly, especially if you already know other languages. How quickly you can understand ML, I don't know, it seems to me to be a very broad term.

[fbrncci]

If all you knew was Rust, then getting a job in this already harsh market would be much harder. It would be easier if you knew ML, but it does really depend on what else you bring to the table. So, what else do you know? The job market might change on a dime, make sure you don't just learn the latest hype cycle of frameworks, languages or try to optimize for hiring right now.

[maara → fbrncci]

I am a backend engineer who have experience building microservices and data pipelines for more than a decade. I felt AI/ML is a natural expansion to my skillset as I currently work on Scala, Apache Spark, Kafka and some Golang.

[fbrncci → maara]

Yes, then I would go the AI/ML route. I'm also mainly a backend engineer. I considered investing time into Rust as well. But employability is questionable. I'm already in the AI/ML field, so I chose to get more experience in the sales & product side of engineering.

[psyklic]

Choose the one you're most excited about learning, then go all in as deep/energetically as possible to stand out. The determining factor isn't the job market; it's yourself becoming demotivated when the going gets tough.

[alexfromapex]

Both will take a while to ramp up to. Rust will have fewer job prospects but machine learning will have much much more competition in the job market.

[nullptr_deref]

Thank you for asking the question.

I am curious to know how long have you even working in tech industry? If you are new, are you optimistic about the future in tech? If yes, how come?

I am getting a feel that we are all transitory generation that will go into postAGI world. Because AGI is already here, at least for programming, as long as they are not hitting the long tail.

Thus my question is, the industry doesn't care about human beings. They treat us as commodity (proved by layoffs). Eventually AI are going to do most of the things. If that is the case, how will people adjust to this?