Close Menu
    Trending
    • MLOps Zoomcamp — 1. I will write directly what I have done… | by Ceyhun Andac, Ph.D. | May, 2025
    • I Learned to Lead by Falling Off a Skateboard
    • Why I’m Excited About Multimodal AI (And You Should Be Too) | by Abduldattijo | May, 2025
    • How to Keep Fatigue From Turning Into Failure
    • Reflections of Artificial Intelligence after reading Mark Levin’s article “Artificial Intelligences: A Bridge Toward Diverse Intelligence and Humanity’s Future” | by Max Thinker | May, 2025
    • AI Models Like ChatGPT Are Politically Biased: Stanford Study
    • My learning to being hired again after a year… Part 2 | by Amy Ma | Data Science Collective | May, 2025
    • Why Gold and Bitcoin Are the Go-To Safe Havens in 2025
    Finance StarGate
    • Home
    • Artificial Intelligence
    • AI Technology
    • Data Science
    • Machine Learning
    • Finance
    • Passive Income
    Finance StarGate
    Home»Machine Learning»My learning to being hired again after a year… Part 2 | by Amy Ma | Data Science Collective | May, 2025
    Machine Learning

    My learning to being hired again after a year… Part 2 | by Amy Ma | Data Science Collective | May, 2025

    FinanceStarGateBy FinanceStarGateMay 18, 2025No Comments19 Mins Read
    Share Facebook Twitter Pinterest LinkedIn Tumblr Reddit Telegram Email
    Share
    Facebook Twitter LinkedIn Pinterest Email


    One Yr Later: What I Realized Nonetheless Issues

    Picture by Nathan Lemon on Unsplash

    That is the second a part of “My Journey to Getting Hired Again After a Year.” Onerous to consider, but it surely’s been a full yr since I revealed the primary half over on TDS. And in that point, one thing stunning occurred. On occasion, somebody would depart a remark, spotlight a sentence, or ship me a message. Most have been easy notes like, “Thanks, Amy. Your submit helped me.” However these phrases lit me up. They brightened total days. They jogged my memory that I used to be by no means really alone, not throughout these lengthy months of unemployment, not within the battle of determining who I used to be and not using a job title or firm identify beneath my e-mail signature or Linkedin profile.

    Humorous sufficient, these onerous days turned out to be a number of the most significant ones I’ve had. Possibly much more significant than my busiest days at work. As a result of in dropping an identification, I discovered new ones. I didn’t want a job or a title to really feel related. To a lot of you, I’m only a fairly lazy author getting again into the groove. And right here I’m — returning to my writing routine. So, thanks to everybody who reached out. Your messages rank second on my record of happiest issues folks give me. The primary? That’s simple. My daughter Ellie’s three S’s: her odor, her smile, and her surprises.

    Sufficient discuss. Let’s get into Half 2. I’ll choose up the place I left off – sharing the teachings that helped me get employed once more. This time, I’ll additionally mirror on how these classes present up in my work and life as we speak. And for these of you interested by the strategies from the guide By no means Search Alone, I’ve acquired some ideas on that too. What labored, what didn’t, and the way I made it my very own.

    A yr into working as a machine studying engineer, I can say this was my largest life lesson.

    Right here’s the backstory. I’d been working as a knowledge scientist ever since I completed grad faculty. Over the previous 7 years, I’d constructed a number of machine studying fashions, linear regression, neural networks and Xgboost. All strong stuff. However when it got here to designing a complete machine studying system from begin to end? That was a special story. I hadn’t actually accomplished that. I knew the way to develop fashions, positive. I even had some expertise deploying them, however solely elements of the method. In case you requested me to design, construct, and run a complete system end-to-end, I couldn’t say I had that have.

    And the job market? It was altering quick. Firms didn’t need somebody who might simply construct fashions anymore. Generative AI was dealing with quite a lot of the info evaluation now. What they actually wished was somebody who might take machine studying and use it to resolve actual enterprise issues, somebody who might personal the entire course of. In the meantime, I had simply been laid off. I had time. So I made a decision perhaps this was the fitting second to pivot. Possibly it was time to go for machine studying engineering.

    The very first thing I did was attain out to individuals who had already made that transfer. Two associates mentioned sure. One had gone from knowledge scientist to machine studying engineer. The opposite was a knowledge scientist, and her husband labored as an MLE at Apple. We ended up having this lengthy cellphone name for 2 hours, perhaps extra. They have been sort. They usually didn’t sugarcoat something. Each of them advised me it was robust to make the swap. Not unattainable, however robust. In case you didn’t have MLOps expertise or a strong GitHub portfolio to indicate off, touchdown a senior MLE job can be actually onerous. Particularly with how aggressive issues have been getting.

    That dialog hit onerous. I bear in mind feeling my coronary heart pound, like chilly water had been poured over my head. I had two choices: I might maintain chasing knowledge scientist jobs — utilized scientist roles at locations like Amazon — however there weren’t many on the market. Or swallow my satisfaction, let go of seven years of expertise as knowledge scientist and go for an entry-level MLE position. Actually, neither selection felt nice.

    It took me two weeks to work by way of it. Two lengthy lengthy weeks. However in the long run, I made up my thoughts: I’d strive for machine studying engineer jobs not less than, even when I needed to begin from the underside. I acquired again to my routine and prepped for interviews. Throughout these onerous days, I began running a blog right here on Medium to indicate my technical muscle, sharing my “Braveness to Be taught ML” collection. Prepared for a spoiler alert? I ended up with three presents for senior and even employees stage machine studying engineering roles. And I had three different final-round interviews lined up that I needed to stroll away from, as a result of there simply wasn’t sufficient time or vitality for me to do all of them.

    No, none of these presents got here from FAANG corporations. However I’m very happy with the place I landed. It was definitely worth the strive.

    Even now, scripting this, I can nonetheless really feel that chill from when my associates advised me the percentages have been slim. And I can nonetheless chuckle at how panicked I used to be. Simply the opposite day, I spoke with a pal who’s trying to transfer from knowledge engineering into MLE. I advised him the identical factor I realized for myself: You are able to do it. And should you determine it’s value making an attempt, don’t get hung up on the percentages. Even when it’s a 1% likelihood, why not see should you’re in that 1%? However should you don’t strive in any respect, you’re 100% within the group that by no means made it.

    For me, the takeaway is straightforward. Don’t be afraid of chances. Even 99.999999% shouldn’t be 100%. In case you’re frightened concerning the consequence, cease serious about the result. Simply do it for enjoyable, on your psychological well being, for the prospect to stay with out regrets.

    A Yr Later: I take advantage of this lesson nearly day-after-day. I weblog shamelessly, pretending I don’t care concerning the quantity followers. I make these awkward customer support calls, simply to see if somebody on the opposite finish may truly assist me. I even purchase a lottery ticket every now and then when the jackpot tops a billion {dollars}. Who is aware of? I would find yourself in that 0.0000…001%. And you recognize what? I just lately gained $12 on a ticket. So sure — it’s value making an attempt.

    This was one other onerous lesson from my “to be an MLE or to not be” chapter.

    Once I spoke with these two associates, they made one factor clear. If I wished to change into a machine studying engineer, I wanted hands-on expertise with MLOps (machine studying operations). The issue? In my previous roles, I’d both handed off my fashions to software program engineers for deployment or dealt with only one small a part of the system myself. I knew I had a spot. And my first intuition was to fill it by any means crucial. So I figured, why not get entangled in some actual initiatives? One thing advanced. One thing I might proudly add to my resume.

    Since I used to be out of labor, I had time. I joined MLOps communities on Slack and Discord. I posted about my background, supplied to work without cost with any startup or crew that wanted assist. Simply to get some expertise in change. The response? Fairly discouraging. Hardly anybody replied. A number of did, however they anticipated me to work 50+ hours per week… without cost and with none working plans. I bear in mind sending a message to a PhD scholar after studying his job posting. I advised him how I appreciated his work and wished to make his product a actuality. He didn’t get again with me. He as an alternative modified his posting to say he was searching for skilled MLEs or somebody with a PhD. Ouch.

    After a number of weeks of all that, I used to be demotivated and burned out. I used to be pleading for alternatives and it was clear. It was then that I made a decision to affix a Job Search Council (JSC) (I defined JSC intimately within the half 1). We shared the emotional weight of job looking each Friday. I slowly began letting go of the strain. And that’s when one thing clicked. I wanted to cease pleading for jobs. As an alternative, I made a decision to promote what I had.

    I rewrote my resume into two variations, one for knowledge scientist roles and the opposite for MLE roles. I utilized for MLE jobs crazily simply to extend the possibilities. However this time round, I approached it in a different way. I broke down what the hiring managers have been truly searching for in an MLE. I noticed how all of the mannequin constructing expertise I had acquired had truly taught me on debugging, monitoring, and resolving messy enterprise issues. Whereas I didn’t have quite a lot of MLOps expertise, I wasn’t coming from zero. I had a grasp’s diploma in pc science, I used to be accustomed to software program growth, and I knew knowledge engineering.

    In these MLE interviews, I began highlighting these abilities. I defined how I utilized machine studying to resolve enterprise issues, supplied delicate hints about favourite model-training tips. I confirmed hiring managers I knew the way it felt to run techniques into manufacturing. I used to be sincere about the place I wanted to realize extra expertise on. However I made it clear this wasn’t a chilly begin.

    Sooner or later, I finished appearing like a job-begger and have become a salesman. I wasn’t asking somebody to “please rent me. I’m keen to work extra and cheaper”. I used to be promoting one thing. When an organization didn’t rent me, it wasn’t a rejection. It simply meant they didn’t want somebody like me. Possibly I wanted to tighten the pitch the following time.

    This psychological shift made all of the distinction. Damaging suggestions wasn’t private anymore. It was simply suggestions, a little bit knowledge level I might use to make changes. Whenever you ask for one thing, folks assume much less of you. However whenever you deal with your self as a product, you’re refining and looking for the fitting consumers. If there’s a flaw, you repair it. If there are good issues, you level them out. And eventually, you discover your folks.

    A Yr Later: I don’t beg anymore. Not for jobs. Not for alternatives. I change. I promote. That mindset has change into a part of me now. It’s my inside tiny salesperson.

    Mock Interviews and the Interview Marathon: Follow Actually Does Make a Distinction

    I’ll be straight with you. Earlier than I began interviewing for machine studying engineer roles after my layoff, I had by no means actually practiced behavioral interviews. Not as soon as in my seven years of working. Certain, I wrote out a number of tales utilizing the STAR technique, like everybody says it is best to. However I by no means practiced them out loud, and I undoubtedly by no means acquired suggestions. It was like stepping on stage to carry out in a play with out ever going to rehearsal. I by no means realized how large a mistake that was, in all probability as a result of, again when the job market was good, I didn’t need to.

    However after the layoff? After spending almost a yr at dwelling due to being pregnant? The market was ice chilly. There weren’t many possibilities, and I couldn’t afford to blow any of them. I had to nail the behavioral interviews. Not simply by memorizing my tales, however by truly practising. For actual.

    So, I made my husband do mock interviews with me. I sat in a single room, he sat in one other, and we jumped on Zoom prefer it was the true factor. Poor man — he’s been on the identical job since ceaselessly and works in a completely totally different discipline, however there he was, asking me random behavioral questions. At first, I didn’t assume it was going to assist. I figured he didn’t get what I did anyway. However once I began answering with my “well-crafted” tales, one thing stunning occurred. I acquired nervous. And wordy. Means too wordy.

    After which he reduce me off. Not gently, both. He advised me straight up: I used to be spending approach an excessive amount of time speaking concerning the background. The corporate, the challenge, all of the setup. He mentioned by the point I acquired to the half about what I truly did, he had already tuned out. what? He was 100% appropriate and I’d by no means seen it earlier than. I by no means thought of how a lot time I used to be losing on particulars that didn’t actually matter to the particular person listening.

    After that, I went again by way of my tales. Nearly all of them had the identical drawback. An excessive amount of setup, not sufficient deal with motion and outcomes. Actually? I used to be grateful for his brutal suggestions. It was a little bit embarrassing, however I wanted I’d accomplished mock interviews like that years in the past.

    From then on, I made a decision to follow much more. With my new MLE resume prepared, I began making use of like loopy. Interviews got here in, and as an alternative of making an attempt to keep away from them, I leaned in. Earlier in my profession, I used to be the sort of one that’d seize the primary supply simply to flee the stress of interviewing. Promoting myself has all the time made me a little bit panicky. In spite of everything, I’m an introvert. However this time, issues have been totally different. The guide By no means Search Alone and people early mock interviews modified my mindset. (I’ll discuss extra concerning the guide and why it prevents me to hurry out of the interview course of later.)

    So I gave myself time. I mentioned sure to nearly each interview I might get. At one level, I interviewed with 4 corporations over three days. It felt like a marathon, however someplace alongside the best way, I acquired good at telling my story. I watched how interviewers reacted. I collected suggestions from the method. And one thing unusual occurred: I finished caring a lot concerning the outcomes. Whether or not I acquired a sure or a no didn’t shake me anymore. I wasn’t simply interviewing to get a job. I used to be practising to get the job I actually wished.

    By the point I had three presents on the desk and at last selected the one I appreciated, I knew I used to be accomplished. That was my end line. It felt like I’d run the complete race and truly gained the prize I wished not the one I settled for.

    Critically, I can’t say this sufficient: KEEP interviewing. Again-to-back should you can. Do mock interviews with whoever you belief, even when they aren’t in your discipline. Follow till you’re much less frightened concerning the consequence and extra centered on getting higher.

    A Yr Later: It’s onerous to say how a lot of these interview abilities I nonetheless have in me now. But when I ever must follow once more, you higher consider I’ll be dragging my husband again into one other spherical of mock interviews. Possibly even for enterprise shows. He’s a troublesome crowd, however he will get outcomes :]

    Panic Mode? Deep Breath , the Present Should Go On

    Throughout my interview marathon, I began noticing one thing that utterly threw me off. Some interviewers regarded… upset. Others appeared bored. And me? I cared. Lots. Most likely an excessive amount of. Each time I noticed a face that wasn’t smiling or nodding, I panicked. In my head, I’d hear this loud voice saying, “Amy, you’re blowing it.” And as soon as that thought crept in, it was over. My mind and physique would scramble to repair the scenario, so I’d begin speaking quicker, throwing out extra phrases, hoping to vary their minds. I wished to come back throughout as sharp and spectacular. However the reality is, I in all probability regarded like a nervous, rambling mess.

    My husband confirmed it after considered one of our mock interviews. He didn’t sugarcoat it. “You’re not even wanting on the digital camera,” he mentioned. “And also you appear actually tense.” Once more, he’s the fitting.

    For an introvert like me, fixing this wasn’t simple. However I discovered two issues that helped. So I’ll share it right here.

    The primary was easy: breathe. Each time I noticed what I thought was a foul response, a frown, a yawn, that clean expression that felt like doom, I pressured myself to pause. I took a breath. And as an alternative of dashing to say extra, I slowed down. Generally I even cracked a chilly joke. (I’m surprisingly good at unhealthy jokes. It is likely to be my secret expertise.) Then I’d apologize for the joke, take one other breath, and transfer on. That little reset labored in two methods. First, it quieted the voice in my head screaming “You’re ruining this!” Secondly, it made the interviewer’s expression change. Possibly they smiled and acquired the joke. Possibly they only regarded confused and didn’t prefer it. However not less than they weren’t bored or upset anymore. I’ll take that.

    The second factor I did was tape an image of my daughter proper behind the digital camera. Her large, shiny smile was proper there, and each time I glanced at it, I smiled too. Which, by the best way, made me look extra relaxed and human on digital camera. Generally the interviewer smiled again, and similar to that, the vitality shifted. I wasn’t panicking anymore. I used to be again in management. The present was again on.

    I began considering of myself as a salesman. Or perhaps a showman. What do they do when the viewers seems to be drained or distracted? They maintain going. They modify. They convey the vitality again. In case you’re like me, somebody who takes these reactions personally, that you must have a plan. These have been my two tips. You’ll in all probability discover your personal. However the level is: don’t panic. Pause. Breathe. Nobody will discover. After which, get again to the present.

    A Yr Later: Actually, this is likely to be a very powerful ability I picked up throughout that robust yr. I nonetheless use it on a regular basis at work. Once I’m presenting my work to a room full of individuals, I gradual myself down. I image myself in a elaborate tailcoat, like an old-school showman, promoting my concepts to the viewers. Generally I throw in considered one of my basic chilly jokes to maintain issues gentle.

    Once I wrap up a presentation, I ensure to provide folks one thing simple to take with them. I’ll say, “In case you’re heading out and wish one factor to recollect about this challenge, right here’s the punchline.” Then I boil it down to at least one or two sentences and say it clearly. Loud sufficient to stay.

    I even use this trick in common conversations particularly the awkward ones. Somewhat pause makes every thing much less uncomfortable. And as a rule, issues end up higher after that second to reset.

    Do the Mnookin Two-Pager train: How I Discovered a Job That Really Match Me

    I maintain mentioning the guide By no means Search Alone, and there’s a cause for that. Once I first heard about it, I used to be skeptical. As an introvert, the concept of becoming a member of a bunch of strangers to speak about job looking made me extraordinarily unsure and nervous.

    My first group didn’t go nicely. There have been 5 of us, however two folks refused to comply with the method. They have been usually late or skipped conferences solely. It was irritating, and I nearly gave up. As an alternative, I discovered one other group by way of the Slack neighborhood. That point, it clicked. We met each Friday, stored one another accountable. We helped each other keep sane by way of the search. It made an enormous distinction. If you wish to know extra about how the JSC (Job Search Council) helped me, I wrote about it in part one of this story.

    Wanting again, one other helpful issues the guide supplied was the Mnookin Two-Pager train. You sit down and write out what you’re keen on in a job, what you hate, and what your profession objectives are. Easy, however surprisingly highly effective. It pressured me to get sincere with myself. With out it, I in all probability would have grabbed the very first supply and rushed out of the market, simply to be accomplished with it. I’ve accomplished that earlier than. And regretted it.

    This time was totally different. My two pager record stored me grounded. I knew what I wished and the place I wasn’t keen to settle. That’s how I ended up at Disney. The position hits about 85% of what I hoped for. Extra importantly, it steers clear of each crimson flag on my “onerous no” record. A yr later, I’m nonetheless glad I took the time to determine precisely what I used to be searching for earlier than saying sure to something.



    Source link

    Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
    Previous ArticleWhy Gold and Bitcoin Are the Go-To Safe Havens in 2025
    Next Article AI Models Like ChatGPT Are Politically Biased: Stanford Study
    FinanceStarGate

    Related Posts

    Machine Learning

    MLOps Zoomcamp — 1. I will write directly what I have done… | by Ceyhun Andac, Ph.D. | May, 2025

    May 18, 2025
    Machine Learning

    Why I’m Excited About Multimodal AI (And You Should Be Too) | by Abduldattijo | May, 2025

    May 18, 2025
    Machine Learning

    Reflections of Artificial Intelligence after reading Mark Levin’s article “Artificial Intelligences: A Bridge Toward Diverse Intelligence and Humanity’s Future” | by Max Thinker | May, 2025

    May 18, 2025
    Add A Comment
    Leave A Reply Cancel Reply

    Top Posts

    Hfhhdhug

    April 2, 2025

    Boost Your Profits and Customer Loyalty With These 6 Business Strategies

    February 2, 2025

    Creating an End-to-End Deep Learning Project with FastAI | by ServerWala InfraNet FZ-LLC | Feb, 2025

    February 3, 2025

    This Is the Underappreciated Marketing Approach That Will Help You Keep Customers Longer

    February 18, 2025

    The Secrets to Success for Alexander’s Patisserie

    May 13, 2025
    Categories
    • AI Technology
    • Artificial Intelligence
    • Data Science
    • Finance
    • Machine Learning
    • Passive Income
    Most Popular

    How I Built 22 Thriving Businesses United by One Powerful Mission

    April 9, 2025

    The Automation Trap: Why Low-Code AI Models Fail When You Scale

    May 17, 2025

    🚀 Explore Generative AI with the Vertex AI Gemini API — My Google Cloud Skill Badge Journey | by Arpit Jain | Apr, 2025

    April 29, 2025
    Our Picks

    How Likely Are You to Be Diagnosed with Cancer? A Data-Driven Approach | by Shirish Ghimire | Feb, 2025

    February 10, 2025

    Virtualization & Containers for Data Science Newbies

    February 12, 2025

    One-Tailed Vs. Two-Tailed Tests | Towards Data Science

    March 6, 2025
    Categories
    • AI Technology
    • Artificial Intelligence
    • Data Science
    • Finance
    • Machine Learning
    • Passive Income
    • Privacy Policy
    • Disclaimer
    • Terms and Conditions
    • About us
    • Contact us
    Copyright © 2025 Financestargate.com All Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.