Within the Writer Highlight collection, TDS Editors chat with members of our neighborhood about their profession path in information science and AI, their writing, and their sources of inspiration. At this time, we’re thrilled to share our dialog with Vyacheslav Efimov.
Vyacheslav is a Senior Machine Studying Engineer specialising in NLP and Laptop Imaginative and prescient. One in all his passions is making instructional content material: Vyacheslav has printed over 60 articles on In direction of Information Science, explaining advanced ideas in easy phrases, offering wealthy visualizations.
You wrote many newbie and explanatory articles on TDS. Has educating the basics modified the way you design or debug actual methods at work?
I discover the correlation between the extra I train one thing, the higher I perceive it. In actual life, after I write a brand new article, I try to dive into small particulars whereas maintaining the reason easy for my readers. Emphasizing data on this manner helps me higher perceive the workflow of algorithms.
In that sense, each time I encounter an error in one of many algorithms used at work on which I wrote an article prior to now, there’s a increased probability that I’ll quickly determine the answer to the issue on my own. From one other perspective, after I write an article on an unfamiliar subject and discover it myself, it will increase my confidence after I apply that individual algorithm at work, as I already know its utility scope, benefits, disadvantages, and particular particulars or constraints.
This fashion, I can give you authentic options that aren’t apparent to others and again up my option to different teammates, managers, or stakeholders. That data is treasured to me.
With so many new fashions popping up daily, it’s straightforward to really feel fully swamped. How do you resolve what’s price a ‘deep dive’ and what you simply ‘get the gist of’? Has your technique for managing this modified in any respect just lately?
At this time, we certainly have an abundance of fashions and instruments that seem daily. It’s straightforward to really feel misplaced whenever you’re uncertain about what to pursue subsequent.
With restricted time, I often delve deeper into matters that could be relevant at work or in my private tasks. This provides me extra confidence when I’ve to current or clarify my outcomes.
Companies often wish to obtain working outcomes as quickly as doable. That is additionally one of many explanation why, in my articles, I focus extra on theoretical ideas, as I can’t dedicate my time at work to going into theoretical depth.
This fashion, I’ve an environment friendly mix of sensible expertise at work and theoretical insights in my weblog. Each of those elements are necessary for expert information scientists.
You’ve competed in AI hackathons. What did you study from having such tight deadlines? Did it power you to get higher at scoping tasks or deciding on a mannequin? And do you end up utilizing any of these ‘hackathon classes’ whenever you’re sketching out a brand new concept from scratch?
Hackathons sometimes final between a number of hours and two days. That may be a very small time frame to develop a totally practical product. Nevertheless, on the similar time, it pushed me quite a bit prior to now to raised prioritize the options on which I ought to focus. Normally, time administration is a beneficial ability to have. When you could have a number of doable options to deal with your drawback, you should select the one that most closely fits the enterprise wants whereas additionally respecting time constraints.
What can also be nice is that after each hackathon, you may consider your self when it comes to the time it took you to implement sure options. For instance, let’s say that it was the primary time you needed to develop a RAG pipeline, which took you round 4 hours to implement. The subsequent time you face a similar drawback at work or a hackathon, you should have a greater estimate upfront of how a lot time it could take in the event you resolve to make use of the identical technique. In that sense, the hackathon expertise lets you higher outline cut-off dates for the strategies you wish to implement in tasks.
For me, the most important lesson from the hackathon was not specializing in perfection when creating the MVP. Whereas an MVP is necessary, additionally it is essential to current your product attractively to shoppers or traders, clarify its enterprise worth, the issue it solves, and why it’s higher than current options in the marketplace. On this regard, hackathons train you to give you higher concepts that remedy actual issues whereas additionally delivery the MVP shortly, containing probably the most important options.
For readers eager about their profession path: your “Roadmap to Becoming a Data Scientist” collection spans fundamentals by means of superior ML. Should you had been rewriting it as we speak, what matters would get promoted, demoted, or lower completely, and why?
I wrote this text collection a yr in the past. For me, all of the ideas and matters I listed are updated for aspiring information scientists. All math, pc science, and machine studying matters I current there are a necessary basis for any machine studying engineer.
As we’re now in late 2025, I might additionally add a requirement to have no less than minimal expertise with immediate engineering and to be accustomed to some AI-generative instruments, akin to GitHub Copilot, Gemini CLI, and Cursor, which might enable for elevated work effectivity.
As a word, in comparison with earlier years, IT corporations have increased necessities and expectations for junior engineers getting into the info science discipline. It is sensible, as fashionable AI instruments can carry out junior-level duties very effectively, and plenty of corporations choose to depend on them now slightly than on entry-level engineers, as they don’t must pay salaries whereas in each circumstances they obtain the identical outcome.
That’s the reason, if a machine studying engineer possesses the sturdy basic abilities I described in that collection of articles, it will likely be a lot simpler for them to dive autonomously into extra advanced matters.
Your background blends software program engineering and ML. How does that basis form the best way you write?
Having sturdy software program engineering abilities is likely one of the finest benefits you may have as a Information Scientist:
- It makes you notice the significance of well-structured software program documentation and creating reproducible ML pipelines.
- You perceive higher how one can make your code clear and readable for others.
- You perceive algorithmic constraints and which information construction to decide on for a specific activity, primarily based on system wants.
- You possibly can extra simply collaborate with backend and DevOps engineers on integrating your code modules.
- You do not want to depend on others to make SQL queries to retrieve crucial information from the database.
The checklist can go on and on…
Talking of my articles, I don’t have many who current quite a lot of code. Nevertheless, each time I do, I try to make it readable and comprehensible to others. I all the time put myself within the footwear of others and ask myself how my article textual content or code examples can be straightforward to understand or reproduce if I had been in others’ footwear. That is the place the software program engineering expertise makes this realization extra important for me, and I comply with the perfect established practices to ship my ultimate product.
Taking a look at your portfolio and GitHub, you’ve blended software program engineering fundamentals with ML from the beginning. What’s one engineering behavior you would like extra aspiring information scientists adopted early?
Many engineers, particularly juniors, are inclined to underestimate the significance of making good documentation and reproducible pipelines. This additionally occurred to me prior to now, after I was extra targeted on growing sturdy fashions or conducting analysis.
Because it turned out, after I needed to change contexts after which a number of weeks later to return to work on the earlier mission, I used to be then spending quite a lot of time determining how one can run my previous code in a messy Jupyter Pocket book or set up crucial libraries once more, the place I might have simply spent a bit of extra time prior to now by growing a well-documented README.md explaining all of the required steps to execute pipelines from zero.
As a result of it was almost unattainable to rerun my pipelines from scratch, I used to be additionally unable to conduct experiments utilizing different entry parameters, which made the state of affairs much more irritating.
It was a painful expertise for me, but one of the vital beneficial classes I’ve discovered. So if I needed to give a chunk of recommendation to an aspiring information scientist on one specific behavior, it could be this:
“At all times make your machine studying pipelines reusable and well-documented”.
Over the previous yr, has AI meaningfully modified how you’re employed each day as an ML Engineer? What bought simpler, what bought more durable, and what stayed the identical?
In recent times, we have now noticed a major rise in highly effective AI engineering instruments:
- LLMs, which may reply to nearly any query, give recommendation, or discover bugs in software program
- Cursor, Lovable, and Bolt are performing as AI-powered IDEs for builders
- AI brokers can full multi-step duties
As a machine studying engineer, it’s important for me to often adapt to those instruments to make use of them effectively.
What turned simpler
Ranging from 2025, I can observe the next constructive affect on my work:
- For me, it turned simpler to quickly check concepts or prototypes. For instance, there have been instances at work after I was given pc imaginative and prescient issues that fell outdoors my space of information. On this manner, I might ask ChatGPT to suggest a number of concepts to unravel them. There have been even instances when ChatGPT generated code for me, and I attempted to execute it with out understanding the way it labored inside.
Then I had two doable circumstances:- If the code ran efficiently and solved the preliminary drawback, then I attempted to go deeper contained in the OpenCV documentation to grasp the way it finally works.
- If the code didn’t remedy my drawback, I might both ignore it, report the error to ChatGPT, or try to seek out the answer myself.
As you may see, I used to be capable of quickly check an answer that would work and save me hours of analysis with none threat.
- One other glorious use case for me was inserting error messages immediately into ChatGPT as an alternative of looking for an answer on the Web. It labored effectively more often than not, however generally it was affected by errors associated to library installations, system errors, and the deployment of pipelines on the Cloud, amongst different points.
- Lastly, I’m a giant fan of AI hackathons! Having instruments that may generate each the frontend and backend of your system makes an enormous distinction for me, as I can now quickly create prototypes and check my MVP in a number of hours. What I develop now throughout one-day hackathons might require a complete week of labor.
What turned more durable / dangerous
- When writing code with AI, there’s a increased risk of delicate information leaks. Think about you could have a file or code fragment containing important credentials that you simply unintentionally feed into an AI mannequin. Then a third-party software will know your delicate credentials. It may possibly occur, particularly in the event you use a software like Cursor and retailer your credentials in one other file slightly than .env. As a consequence, it’s all the time essential to be very cautious.
- One other threat shouldn’t be correctly testing the AI-generated code and never realizing how one can make a rollback. An AI software can introduce invisible errors within the code, notably when it’s used to change or refactor current code. To make sure that AI-generated code doesn’t degrade, it’s essential to totally overview the generated code components, check them, and save modifications in a manner that lets you all the time rollback to a earlier, appropriate model if crucial.
- When relying too closely on generative AI instruments, there’s a threat that the code will grow to be unreadable, comprise excessively lengthy features, exhibit repetition, or stop to perform appropriately. That’s the reason it’s important to grasp that AI instruments work extra successfully on prototyping than on sustaining high-quality manufacturing code.
What remained the identical
What stays fixed for me is the significance of understanding the interior workflow of the algorithms I exploit, sustaining sturdy pc science foundations, and writing high-quality code, amongst different key abilities. In different phrases, the essential ideas of software program growth will all the time be essential to effectively use AI instruments.
In that sense, I like evaluating a set of obtainable AI instruments to an alternative to a junior developer in my workforce, to whom I can delegate much less important duties. I can ask it no matter I would like, however I can’t be 100% positive it can do my duties appropriately, and that is the place the significance of getting sturdy basic experience comes into play.
To study extra about Vyacheslav‘s work and keep up-to-date along with his newest articles, you may comply with him on TDS or LinkedIn.
