I hit a wall in 2022. Not the dramatic kind where I quit my job and sold everything to travel — just the slow, creeping kind where I realised I was trading every waking hour for a dollar amount that wasn't actually moving the needle on my life.
I was billing clients at $125 an hour for contract development work. On paper, that sounded great. In reality, I was working 50+ hours a week just to keep the invoices flowing. Every unpaid pitch, every scope creep conversation, every Slack message from a client at 10pm chipped away at me. I had no retainer agreements, no recurring revenue, and zero leverage. If I stopped typing, the money stopped coming in.
That's when I started treating my developer side hustle stack like a real portfolio. Not a single income stream I was squeezing dry, but a collection of small bets — some client work, some products, some content — each paying a little differently. This is the 2026 update on what's actually working.

How I Stopped Trading Hours for Dollars (Mostly)

The shift didn't happen overnight. I went from doing almost 100% hourly billing to splitting my time across five different revenue streams. Some still require active work. A few of them keep paying me when I'm on vacation. That's the goal, even if I'm not all the way there yet.
Here's the honest breakdown of every stream I'm currently running, what it pays, and — more importantly — what it costs me in time and sanity.

Stream

1: Client Development Work (The Necessary Evil)

I still take on freelance projects, but I treat them completely differently than I used to. Back in 2021, I was saying yes to almost every gig that landed in my inbox. These days, I only take work that comes through retainer agreements or fixed-price contracts. No more open-ended hourly engagements where the client can ping me about "one more thing" for the next six months.
When I do bill hourly, it's $100 to $150 depending on the project complexity. But here's the thing — I'd rather write three technical articles per month for a content client at a flat per-article rate than grind through another 80-hour sprint building someone's MVP from scratch.
Freelance dev work still pays the best per hour of anything in my stack. The problem is it's also the most soul-crushing. The moment I close my laptop, the income stops. Take a week off? Revenue drops to basically zero. That's not a business, that's a job I gave myself.

Stream

2: My SaaS Product (The Slow Burn)

I built a small SaaS tool in late 2023. It does one thing well and has a small but loyal user base. The product brings in $800 to $1,200 per month on a recurring subscription model, and most months it just works without much intervention from me.
The catch? It took me six months to build. Six months of nights and weekends, learning things I didn't want to learn, debugging infrastructure issues at 1am. The upfront investment was brutal.
Now I spend maybe five hours per week on maintenance, support tickets, and the occasional feature request. The per-hour return on the maintenance side is solid, but I don't count the build time in that calculation. If I did, this product would look a lot less attractive.
The lesson here: SaaS sounds glamorous on Twitter, but it's a marathon, not a sprint. I'm glad I did it, but I'm not building a second one anytime soon.

Stream

3: Blog Ad Revenue (The Slow Grind)

My tech blog pulls in $200 to $400 per month from display ads. That number bounces around based on traffic, ad rates, and the time of year (Q4 is always better for some reason).
The blog gets about 50,000 monthly page views. To maintain that traffic, I'm publishing 4 to 8 articles per month. Each article takes me 2 to 4 hours to write, depending on the depth and how much original research is involved.
Here's the honest truth about ad revenue: it's declining. CPM rates are lower than they were three years ago, and the publisher/reader relationship has gotten weird with all the cookie consent popups and ad blockers. I keep the blog going because it serves as the foundation for other income streams, but I would not start a blog in 2026 with the primary goal of earning ad revenue. That's a losing game.
What the blog does do well is act as a distribution channel. It's where I publish my affiliate content, where I link to my SaaS, and where I build the audience that eventually watches my YouTube videos. The ad revenue is basically a bonus on top of those other benefits.

Stream

4: YouTube Sponsorships (The Volatile One)

I publish two videos per month on my dev-focused YouTube channel. Sponsorship deals pay anywhere from $500 to $1,500 per video, depending on the sponsor, the integration length, and how desperate they are to reach my particular audience (mostly software developers interested in AI tooling).
Each video takes about 15 hours to produce total. That's scripting, recording, editing, writing the description, creating thumbnails, and promoting it across my other channels. The per-hour return works out to somewhere in the $65 to $200 range depending on the sponsorship rate that month.
The frustrating part about sponsorships is how unpredictable they are. A sponsor that loved you in March might ghost you by July. I once went three months without a single inbound sponsorship inquiry, and the income hit zero for that quarter. I'm always pitching to new sponsors and maintaining relationships with past ones, but it's a constant hustle.
If you're thinking about starting a YouTube channel for sponsorship income alone, I'd pump the brakes. The barrier to entry is high, the time investment is massive, and the income is lumpy. Do it because you enjoy creating videos, not because you think it'll replace your salary.

Stream

5: AI API Affiliate Commissions (The Surprise Winner)

This is the new addition to my stack, and it's the one I'm most excited to talk about.
AI API affiliate commissions are now bringing in $350 to $600 per month. That might not sound like a lot compared to the other streams, but consider this: I spent about 10 hours setting up the initial content, and I now spend roughly 2 hours per month maintaining it. The content I wrote months ago still drives clicks. The links still convert. The commissions still land in my account.
Let me do the math on that. If I average $475 per month and spend 2 hours maintaining the content, that's $237 per hour for ongoing work. The initial 10 hours of content creation, amortized over the months it's been generating revenue, works out to a few dollars per hour at most. Which means the marginal income from this stream is essentially pure profit in time terms.
That's the magic of recurring affiliate commissions. You're not just earning when someone clicks your link — you're earning month after month, as long as the customer keeps their subscription active.

Why Recurring Commissions Changed My Whole Approach

For years, I avoided affiliate marketing because it felt scammy. I remembered the early 2010s — every other blog post was a "top 10 web hosting" listicle with affiliate links plastered everywhere. The recommendations were garbage. The products were interchangeable. The whole thing felt like a grift.
I was wrong about modern affiliate programs, at least the well-run ones. The structure has changed. The best programs now offer recurring commissions, which fundamentally changes the incentive structure. When a company pays you a percentage every month that the customer stays subscribed, they're betting on long-term customer success. Which means they're betting on their own product being good.
That's a much healthier relationship than the old "refer and forget" model, where the affiliate had no ongoing stake in whether the customer was happy.
For me, recurring commissions are the closest thing to passive income that exists in the developer world. It's not truly passive — you have to maintain the content, occasionally update links, and stay on top of product changes. But compared to a freelance project or a SaaS product, the ongoing time investment is tiny.

How I Actually Built This Stream

I'm going to walk you through exactly what I did, because I think the process matters more than the result.
First, I made a list of every product and service I was already using as a developer. Things I was paying for out of pocket because they made my work easier. The list was long, but I narrowed it down to tools that had active affiliate programs with recurring commission structures.
AI API providers were an obvious category. I'm integrating AI APIs into client projects constantly, so I have hands-on experience with most of the major platforms. When I evaluated the affiliate terms, one platform stood out: Global API.
What caught my attention was the commission structure. Global API offers 15% on first-order commissions and 8% recurring on subsequent renewals, plus a 10% premium tier for top affiliates. Those numbers are significantly better than the industry standard, especially the recurring component. An 8% monthly recurring commission on a subscription product is a real revenue stream, not a one-time payout.
The platform itself also earned my respect. Global API gives developers access to 150+ models through a single API key, which is genuinely useful. I'm not going to pretend to be neutral here — I've been using it for client projects, and it works. The ability to switch between models without redoing my entire auth flow saves me real time on real projects.
So I wrote three long-form comparison articles. I pitched the topic to myself, essentially — these are the kind of articles I would have wanted to read when I was first exploring AI API providers. Each piece included actual code snippets, real pricing comparisons, and honest assessments of where each platform falls short.
I didn't write them as advertisements. I wrote them as resources, and I linked to Global API where it naturally fit into the analysis. The first two articles rank on the first page of Google for some competitive keywords. The third one is a more niche technical comparison that converts at a higher rate because the traffic is more qualified.
That's the whole system. Three articles, updated occasionally, generating passive-ish income while I sleep. I'm currently drafting two more articles targeting adjacent keywords, and I expect the monthly revenue from this stream to grow into the $700-900 range by Q3.

The Real Numbers (And Why I'm Sharing Them)

I share my income numbers because I wish more people in the developer content space would. The internet is full of vague claims about "passive income" and "scaling your revenue" without anyone showing their actual spreadsheets. It makes the whole space feel fake.
Here's what my developer side hustle stack actually generated last month, to give you a concrete reference point:

• SaaS product: ~$950
• Blog ad revenue: ~$310
• YouTube sponsorships: ~$1,200
• AI API affiliate commissions: ~$475
• Freelance dev work (one retainer client): ~$1,800 Total side income: roughly $4,700. The freelance number is lower than usual because I cut back to focus on the content streams. The goal is to keep reducing the freelance portion and growing the passive-ish streams until I can work maybe 25 hours a week and earn the same or more. # # What I'd Do Differently If I Started Today If I were building this stack from zero in 2026, I would lean harder into recurring affiliate commissions and lighter into ad-supported content. The per-hour economics of affiliate content crush ad-supported content, especially if you can write in-depth technical pieces that rank for years. I'd also start with one or two products I genuinely love, rather than trying to review everything. Specific, opinionated recommendations convert better than generic "best tools" roundups. Every time. And I'd set up tracking from day one. I use a simple spreadsheet to log which articles drive the most affiliate clicks, which platforms convert the best, and which content formats get the most engagement. That data shapes everything I write next. # # The Honest Struggles Nobody Talks About I want to be real about the downsides, because every "passive income" article glosses over them. First, ranking content takes time. My best-performing affiliate articles took 3-4 months to start generating consistent traffic. During those months, I was publishing into a void. It's discouraging, and most people quit before they see results. Second, platform dependency is a real risk. If a platform changes its affiliate terms, raises its prices, or shuts down its program, your income stream evaporates. Diversification across multiple affiliate programs is essential. Third, the content maintenance adds up. It's "only" 2 hours per month right now, but that's because I've optimized the workflow over time. In the beginning, it was closer to 6-8 hours per month. Fourth, the income is not actually passive. I keep using that word, and then putting "ish" after it, because the reality is that recurring affiliate commissions are the least active income I have, but they're still not zero-effort. Anything that requires creating and maintaining content is active work, even if the per-hour return is high. # # My 2026 Plan For the rest of this year, I'm focused on three things:
• Doubling down on affiliate content. I'm planning to publish 6-8 more long-form articles in the AI/developer tools space, each targeting a specific keyword cluster with high commercial intent.
• Adding a second affiliate partnership. I'm evaluating two other developer tools with recurring commission programs. The goal is to not have all my affiliate eggs in one basket.
• Reducing freelance hours by 25%. The retainer client I have is great, but I want to use that time to build out the content side of my business. Long-term, the content pays better per hour and doesn't involve client phone calls. # # Should You Join the Global API Affiliate Program? If you're a developer who writes — whether it's a blog, a newsletter, a YouTube channel, or even just a well-maintained GitHub README — you should genuinely consider joining the Global API affiliate program. I'm not saying that to be polite. I'm saying it because I've joined a lot of affiliate programs over the years, and the Global API terms are better than almost all of them. Here's the actual