I Switched Careers at 28 With No Connections. Here's Exactly How.

Published on BirJob.com · March 2026 · by Ismat

On my 28th birthday I was working a job I'd had for four years, making decent money, and feeling like I was slowly suffocating. That sounds dramatic. It was dramatic. I was an economics graduate working in operations at a logistics company in Baku, and every morning I'd sit in traffic on the Heydar Aliyev highway thinking "I can't do this for thirty more years."

I didn't know anyone in tech. My family didn't have connections (my father's a retired teacher, my mother works in a hospital). I didn't have a CS degree. I didn't even have a portfolio. What I had was a laptop, a stubborn refusal to keep being miserable, and exactly 11 months of savings — about 14,000 AZN — that I'd been hoarding because I'm constitutionally incapable of spending money I don't have to.

Two years later, I'd built BirJob, a job aggregator that scrapes 82 websites daily and serves thousands of job seekers. I went from zero programming knowledge to running a production system that processes 9,400+ listings three times a day.

I'm not telling you this to brag. I'm telling you this because when I was at the beginning, I couldn't find a single honest account from someone in the post-Soviet space who'd actually done this. All the career-switch stories were from Americans who went to a $20,000 bootcamp and got hired at Google. That wasn't my reality. This is for people whose reality looks more like mine.

Month 0: The Decision (September 2023)

I didn't quit my job on day one. I'm not insane.

I spent September doing research. What field? What skills? What's realistic? I knew I wanted to work with computers, but "working with computers" is like saying "I want to work with words" — it could mean novelist, journalist, copywriter, translator, or spam email author.

I narrowed it down to three options:

• Data analytics — because my economics background gave me some statistical intuition
• Web development — because it seemed to have the most job openings
• Product management — because people kept telling me my "business background" was relevant (it wasn't, really)

I eliminated product management quickly. Every PM job posting I found required 3+ years of experience in tech. How do you get experience in tech without being in tech? Circular. Gone.

I chose data analytics initially, then pivoted to web development within two months. More on why below.

If you're considering data analytics as a career path, we have a comprehensive data analyst roadmap that covers what you actually need to learn and in what order.

Months 1-3: Learning the Hard Way (October-December 2023)

I started with Python because everyone said start with Python. I used freeCodeCamp, Kaggle's free courses, and YouTube. I avoided paid courses entirely during this phase — not because I'm cheap (though I am), but because I didn't trust myself to know which paid course was worth it yet.

The first month was painful. I remember spending four hours debugging a for loop that didn't work because I'd used a colon instead of a semicolon. Wait, no — Python doesn't use semicolons. It was an indentation error. Four hours on whitespace. I genuinely considered quitting.

I didn't quit. But I did something arguably worse: I tried to learn everything at once. Python. SQL. Pandas. NumPy. Matplotlib. Seaborn. Scikit-learn. Statistics. I was opening a new course every week without finishing the previous one. By December, I "knew" eight different tools and couldn't actually build anything with any of them.

This is the mistake almost every career switcher makes. I call it "tutorial purgatory" and I was trapped in it hard.

The Pivot (January 2024)

Three things happened in January that changed everything.

First, I realized that data analytics jobs in Baku mostly require 1C and Excel. Not Python. Not R. Not the fancy stuff I'd been learning. I was studying for jobs that barely existed locally.

Second, I was manually checking 15 job sites every day for my own job search and thought: "This is stupid. There should be one place that has everything." That thought became BirJob.

Third — and this is the important one — I realized that building an actual project would teach me more than any course. And I had a project idea that solved a real problem. My problem.

So I pivoted. Instead of "learning to be a data analyst," I started "building a job aggregator." The learning became a side effect of the building.

Months 4-6: Building the First Version (January-March 2024)

The first version of BirJob was embarrassing.

I'm talking a Python script that scraped three websites, dumped the results into a CSV file, and a static HTML page that I manually updated twice a week. No database. No automation. No frontend framework. Just raw HTML with some Bootstrap CSS.

Here's what I learned in these three months:

• Web scraping with BeautifulSoup and requests
• Basic HTML/CSS
• How to deploy a static site (I used Netlify initially)
• Git (badly)
• How to read documentation (the actual skill nobody teaches)

I wasn't learning from courses anymore. I was learning by hitting walls. "I need to scrape this page" → google "python web scraping" → find BeautifulSoup → follow the quickstart → hit an error → Stack Overflow → fix it → hit the next error. That's the real learning loop.

Was it efficient? God, no. I spent a whole weekend learning that some websites return different HTML when you don't set a User-Agent header. A tutorial would have told me that in 30 seconds. But I understood it because I'd fought through it.

Month 5: I Quit My Job (February 2024)

This was either brave or stupid. Probably both.

By February, I was working on BirJob from 7 PM to 1 AM every night, plus weekends. I was exhausted. My performance at my day job was slipping. My manager noticed. I was doing two things badly instead of one thing well.

I looked at my savings: 12,000 AZN left (I'd been spending a bit more than planned). That's roughly 8 months of living expenses if I was careful. I did the math. If I went full-time on learning and building, I could either have something viable in 6 months or know with certainty that it wasn't going to work.

I quit. My parents thought I was having a breakdown. My friends thought I was crazy. My former manager actually wished me well, which I'll always respect him for.

Months 6-9: The Real Work (March-June 2024)

Going full-time changed everything. I was coding 8-10 hours a day. I learned:

• PostgreSQL (because CSV files don't scale)
• Next.js (because I needed a real frontend)
• Prisma ORM (because writing raw SQL for every query was killing me)
• Async Python (because scraping 30+ sites synchronously took forever)
• Docker (because "it works on my machine" stopped being acceptable)
• GitHub Actions (because I needed automation)

By June, BirJob was scraping 47 sites, had a proper database, a real website with search and filtering, and was running automatically three times a day. It was ugly. The code was a mess. But it worked.

Honestly? The code is still kind of a mess. Production code at startups always is. You refactor when you have time, and you never have time.

The No-Connections Problem (And How I Solved It)

This is the part most career-switch articles gloss over.

I didn't know anyone in tech. Not one person. In Azerbaijan, where connections (tanishlik) influence everything from getting a job to getting a table at a restaurant, this was a real problem.

Here's what I actually did:

1. I started sharing my progress on LinkedIn. Not humble-bragging. Genuine posts about what I was building and what I was struggling with. "Today I spent 6 hours figuring out why my scraper was getting blocked by Cloudflare. Still haven't figured it out." These posts got way more engagement than polished announcements because people relate to struggle.

2. I went to every tech event in Baku. Azerbaijan has a small but active tech community. Bakutel, local meetups at the ADA University campus, startup weekends. I went to all of them. At first I just listened. Then I started asking questions. Then people started recognizing me.

3. I emailed people directly. Not form letters. Personal emails to specific people whose work I admired. "I saw your talk at [event]. I'm building [thing]. I have a specific question about [topic]." About 1 in 4 responded. That's a good rate for cold emails.

4. I contributed to the community before asking for anything. I shared BirJob freely. I helped a few junior developers debug their code (even though I was barely junior myself). I answered questions in Telegram groups. You build social capital by giving first.

Within six months, I went from knowing zero people in tech to having maybe 20-30 genuine connections. Not LinkedIn connections — people who'd actually reply if I messaged them. That network was built entirely from scratch, one conversation at a time.

Months 10-14: From Project to Product (July-November 2024)

The second half of 2024 was about turning BirJob from a hobby project into something real. I added:

• User accounts and saved searches
• An HR dashboard where companies could post jobs
• Email notifications for new job matches
• SEO optimization (because what good is a website nobody finds?)
• More scrapers, constantly — we went from 47 to 82 active sources

The data analyst roadmap we published became one of our most-read pages, which taught me something important: content is distribution. Writing useful things brings people to your platform.

I also made some spectacularly bad decisions during this period. I spent two weeks building a feature nobody used. I tried to add a chat system and realized I was solving a problem that didn't exist. I deployed a broken update at 2 AM and didn't notice until morning. Classic founder mistakes.

The Money Question

Everyone wants to know about money. Here's the honest version.

For the first 10 months, BirJob made zero revenue. Nothing. I was living on savings, which were shrinking uncomfortably fast. By August 2024, I had about 3,000 AZN left. I was genuinely scared.

I started monetizing in September 2024 by offering paid job posting for companies. Not a big launch — just quietly adding a "Post a Job" button and charging 15 AZN per listing. The first paying customer was a small IT company in Baku. I remember refreshing my bank app every five minutes the day they paid.

Revenue grew slowly. Very slowly. By the end of 2024 I was making about 400-600 AZN per month. Not enough to live on comfortably, but enough that the trajectory was moving in the right direction.

Today? It's better. Not Silicon Valley money, but it's a real income from a real product that I built from nothing. I won't share exact numbers because I'm a private person about finances, but I will say: I'm no longer checking my bank account anxiously.

What I'd Do Differently

If I had to do this again, here's what I'd change:

1. Skip the course hopping. Those first three months of tutorial purgatory were mostly wasted. I should have started building something — anything — by week four.

2. Don't quit your job so early. I quit at month 5. I should have waited until month 8 or 9. The financial pressure was motivating in some ways but counterproductive in others. I made rushed decisions because I was worried about running out of money.

3. Network earlier. I didn't start going to events until month 6. Should have started in month 1. Connections compound over time — the earlier you start, the more value they accumulate.

4. Learn Git properly from day one. I used Git as a save button for six months before learning branches, merging, and pull requests. My commit history from early 2024 is a crime scene. "fix stuff," "update," "please work," "AAAA" — actual commit messages I wrote.

5. Take care of my health. I gained 8 kg in the first six months. Sitting 10 hours a day, eating whatever was fastest, not exercising. By the time I realized what was happening, I felt terrible physically. A career switch is a marathon, not a sprint, and you can't run a marathon if you're not taking care of your body.

Advice for People Considering the Switch

Look. I'm not going to give you the standard motivational speech. Career switching is hard. It's scary. It's lonely. You'll doubt yourself constantly. People around you will doubt you too.

But here's what I know for sure:

You don't need a CS degree. I don't have one. Never will. Nobody has ever asked me for one. They ask what I've built.

You don't need connections to start. You need connections eventually, but you build them along the way. They're a byproduct of showing up consistently, not a prerequisite.

You don't need to be young. I was 28. I've met people who switched at 35, 40, even 45. 28 felt ancient to me at the time, but looking back, it was fine. Age is a perceived barrier, not a real one.

You DO need savings. Or a spouse with income. Or a part-time job. Something. The "just quit and figure it out" advice is irresponsible unless you have a financial cushion. I had 11 months of savings and nearly ran out. If I'd had less, I might have had to go back to my old career before BirJob became viable.

You DO need to build something real. Not follow tutorials. Build. Something that solves a problem. Something you can show people. That project will be your resume, your portfolio, your talking point in interviews, and possibly your livelihood. Tutorials are training wheels. At some point you have to take them off and actually ride.

You DO need to be okay with being bad at things. For at least six months. Maybe a year. You'll write terrible code. You'll make obvious mistakes. You'll feel like an imposter surrounded by 22-year-olds who seem to know everything. That's normal. The only people who never felt like impostors are the ones who never tried anything new.

Where I Am Now

It's March 2026. I'm 30. BirJob scrapes 82 active sources, serves thousands of users, and makes enough money that I can keep doing this. I write code every day. I'm building features I'm genuinely excited about. And when I drive on the Heydar Aliyev highway now (in lighter traffic, because I set my own schedule), I don't feel like I'm suffocating.

Was it worth it? Unambiguously yes.

Was it easy? Unambiguously no.

Would I do it again? In a heartbeat. But I'd bring a bigger savings cushion.

Sources & Resources

• freeCodeCamp.org — where I started learning Python (free)
• Kaggle.com — free data science courses and datasets
• The Odin Project — free full-stack curriculum I used later
• BirJob.com data on job posting trends and skills demand
• State Statistical Committee of Azerbaijan, employment data 2024-2025

I'm Ismat, and I build BirJob — a job aggregator that scrapes 80+ Azerbaijani job sites so you don't have to. If this helped, check our blog for more.