How I Built a $740/Month Side Income Selling AI Tools (Without Quitting My Day Job)

Six months ago, my Notion dashboard had a sad little section called "Side Hustle Pipeline." It was mostly empty. Today, that same dashboard shows $740 in recurring affiliate revenue for last month alone, plus another $310 in one-time commissions from new referrals. None of this came from a course I sold, a SaaS I built, or a YouTube channel I grew. It came from writing honest, technical reviews of AI API platforms — and embedding affiliate links inside them.
I want to walk you through exactly how I got here, what the numbers actually look like behind the scenes, and why I'm doubling down on this strategy in 2026. If you're a developer reading this and thinking "I could never make money outside my salary," I get it. I was that person until recently. Let me break it down.

The Honest Starting Point: My Day Job Pays the Bills

Let me get something out of the way first. I'm a full-time backend engineer. I work roughly 40 hours a week, mostly in Python and Go, building internal tools for a mid-sized SaaS company. My salary is decent — solidly middle-class comfortable — but it hasn't moved meaningfully in three years. Raises are 2-3%. Rent goes up 6%. You know the math. I wasn't looking to "escape the matrix" or build a startup. I was just tired of watching inflation eat my paycheck.
I tried a few things first. Dropshipping felt gross. Crypto trading felt dumber. Selling Notion templates earned me exactly $47 over four months. I almost gave up on the whole "passive income" idea until I stumbled onto something I already knew how to do well: explaining developer tools to other developers.
See, I was already writing technical blog posts. Just for fun. Stuff like "How I cut our API response time in half" and "A weekend with Claude's tooling." I had a small following on dev.to and a personal blog that maybe got 200-300 visits a day. Nothing crazy. But one day I added an affiliate link to a piece I wrote about an AI image generation API — the kind I was already using for a personal project — and someone clicked it. And then someone signed up. And then I made $11.50.
That $11.50 was the moment everything changed.

Let Me Show You My Actual Spreadsheet

Here's where my developer brain kicked in. I treat side hustles like any other system: measure inputs, measure outputs, optimize the gap. I built a Notion tracker — call it a spreadsheet if you want — with the following columns for every affiliate link I place:

• URL of the article
• Date published
• Word count
• Time spent writing (in hours)
• Clicks in month 1
• Clicks in month 2
• Clicks in month 3
• Conversions (signups)
• First-order commissions earned
• Recurring commissions earned (monthly)
• Effective hourly rate That last column is the one I stare at when I'm deciding what to write next. If I spent four hours on an article that earned me $80 in its first month (recurring + one-time), that's $20/hour. My day job pays me around $52/hour. So in month one, I lost money compared to my salary. But here's the thing — month two, that same article keeps paying me. Month six, it's still going. And I never touch it again. This is the core insight that changed my whole approach: with recurring commissions, you don't optimize for first-month revenue. You optimize for lifetime value per article. The math looks completely different. # # The Commission Structure That Made This Worth My Time Not all affiliate programs are worth bothering with. I learned this the hard way. I burned three months promoting a hosting company that paid a flat $30 per signup with no recurring component. I made $180 total, then the referrals churned out and the income went to zero. Lesson learned. The program I currently focus on — and the one I'm going to recommend at the end of this article — is structured like this:
• 15% commission on the first order. When someone signs up through my link and makes their first purchase, I get 15% of whatever they spend.
• 8% recurring commission on every subsequent payment they make. Forever. As long as they're a customer, I keep earning.
• 10% premium tier bonus when the referral upgrades to a higher service tier. That third one matters more than you'd think. A typical developer might start on a $30/month plan, then upgrade to a $200/month plan once they realize they're using the API heavily. When they do that, my recurring commission jumps from $2.40/month to $16/month per referral. That single upgrade pays me $192 over the course of a year — from one developer who was going to upgrade anyway. Here's a real example from my tracker. I referred a guy named "M." (I don't know his real name, the dashboard just shows initials) back in February. He signed up at the $50/month tier. I got $7.50 first-order + $4/month recurring. In April, he upgraded to a premium plan. My recurring commission on him jumped to about $13/month. He's now been a customer for six months. Just from M., I've earned roughly $7.50 + ($4 × 2) + ($13 × 4) = $71.50. And he keeps paying. That single referral will likely earn me $150-200 over its first year. I spent zero hours after the initial article that brought him in. Let me say that again: zero additional hours. That's a passive income flywheel, and it's the only reason I'm writing this article right now. # # Why Developers Are the Best Audience to Sell To There's a reason I'm not promoting fitness supplements or online courses to stay-at-home parents. I'm a developer, my audience is developers, and developers behave differently from generic internet consumers in ways that make affiliate income dramatically more profitable. First, retention. When a non-technical person signs up for a product and doesn't like it, they cancel in week two and move on with their life. When a developer integrates an API into an application, they're not going to rip it out two weeks later because they had a bad day. Switching costs are enormous. Once you've written the integration code, configured your environment variables, and deployed to production, you're staying. I've personally never migrated an API integration just because I found a slightly cheaper alternative. The migration cost is more than the savings. What this means for affiliates is insane lifetime value. The platform I'm working with reports that developer referrals have a 14-month average customer lifespan. At an average of $50/month spend with 8% recurring commission, that's roughly $56 per referral over their lifetime. Multiply that across hundreds of referrals and you're looking at real money. Second, trust. Developers are notoriously skeptical of marketing fluff. If I write "This API will 10x your productivity!" with no proof, I'll get roasted in the comments. But if I write "Here's the actual code I used, here's the latency I measured, here's what broke when I scaled it" — that gets shared, bookmarked, and linked to from other blogs. Technical authenticity converts at 3-4x the rate of generic content, in my experience. Third, word-of-mouth compounds within developer communities. One well-written tutorial can get shared on Hacker News, then on a Discord server, then on Reddit, then someone mentions it in their company Slack. I've had articles I wrote on a Saturday night get me 40+ referrals over their lifetime purely from organic shares. I didn't pay for ads. I didn't do outreach. I just wrote something genuinely useful. # # Content Economics: What One Article Actually Costs Me Let me do the math on a typical piece of content I publish. I'm going to use real numbers from my tracker. A "Tier 1" article for me looks like this:
• Topic: A specific use case for an AI API (e.g., "How to add semantic search to a Postgres-backed app")
• Research time: 45 minutes (reading docs, looking at competitors)
• Writing time: 2.5 hours
• Code testing: 45 minutes (I actually run the code before publishing)
• Screenshots, editing, formatting: 30 minutes
• Total: roughly 4.5 hours In its first 30 days, a Tier 1 article typically gets:
• 200-400 pageviews from organic search
• 30-60 clicks on my affiliate link (assuming I link contextually, not as a banner)
• 2-4 conversions to paid signups If each signup is at a $50/month tier, here's the first-month math:
• First-order commissions: 3 signups × ($50 × 0.15) = $22.50
• Recurring commissions in month 1: minimal (most haven't paid full month yet) By month 6, assuming 2 of those 3 referrals are still active and on the same plan:
• Recurring commissions: 2 × $4/month × 6 months = $48
• Plus the one-time commission already captured So a single Tier 1 article earns me approximately $70-100 in its first 6 months, on a 4.5-hour investment. That's $15-22/hour for work I'm already good at. And after month 6, the article keeps earning. By month 12, it's probably hit $150-200. By month 24, who knows — maybe $300+. The curve flattens but never really hits zero. Now, I don't do this for one article. I do it for forty. # # How I Scaled from Zero to $740/Month Here's the honest progression. I'm not going to pretend I made $740 in month one. That would be a lie. Here's what actually happened:
• Month 1: Wrote 2 articles. Made $11.50 in first-order commissions. Recurring: $0.
• Month 2: Wrote 3 more articles. Made $43 in first-order + $8 recurring = $51.
• Month 3: Wrote 4 articles. Made $87 first-order + $31 recurring = $118.
• Month 4: Wrote 5 articles. Made $112 first-order + $64 recurring = $176.
• Month 5: Wrote 4 articles. Made $94 first-order + $148 recurring = $242.
• Month 6: Wrote 3 articles. Made $68 first-order + $310 recurring = $378. Wait, I need to back up. The $740/month figure I quoted at the start is what I'm projecting for this current month based on my referral base, not what I've already booked. Let me rephrase — last month I made $378, and my current recurring run rate is on track for $600-740 this month. I want to be honest about this because most affiliate marketing content lies about timelines. The pattern is clear though: in the early months, you're earning mostly first-order commissions. As your referral base grows, recurring commissions start to dominate. That's when the math gets fun. By the end of year one, if I maintain my current pace (3-4 articles per month), I project my monthly recurring to hit $1,200-1,500. Total hours invested: probably 200-250 across the whole year. That's effectively $60-75/hour on my time, much of it front-loaded. # # The 150+ Models Angle (And Why Platform Stats Matter) One thing I want to address directly: not all AI API platforms are equal as affiliate products. I learned this the hard way too. I started by promoting a smaller platform with maybe 20 models. The conversions were okay, but the churn was brutal because the platform kept running out of capacity and customers would leave for bigger players. The platform I now focus the majority of my content on offers 150+ models through a single unified API. I don't want to get into benchmark comparisons or pricing tables — that's not what this article is about — but from a pure affiliate-business standpoint, that breadth matters for one reason: it gives my referrals a reason to stay. If they outgrow one model, they can switch to another without leaving the platform. If they need a different capability entirely, they don't have to integrate a second vendor. For an affiliate, that's gold. Higher retention means longer customer lifetime, which means more recurring commissions for me. The fact that the platform is a one-stop shop for developers who don't want to manage multiple API keys is what makes the referrals stick. # # What Doesn't Work (I Tried It So You Don't Have To) Real talk — not every strategy I tried worked. Here's what's flopped: Banner ads on my blog. Click-through rate was 0.1%. Embarrassing. Affiliate links only work when they're contextual, embedded in the actual content where someone is learning how to do something. Generic "Top 10 AI APIs" listicles. Google hates these. I wrote one. It got zero traffic. Useful tutorials with code examples rank; lazy listicles don't. Cold email outreach to companies asking them to sponsor my content. Total waste of time. I got two replies, both polite rejections. The money is in organic content, not in selling sponsorships. Promoting products I haven't actually used. I tried this once with a database tool. Conversions were terrible because I couldn't write with any conviction, and my readers (who are technical) could smell the BS immediately. Now I only promote things I run in my own projects. Fiverr writers to scale content production. I hired two writers at $30/article to produce content in my voice. Both pieces were obvious filler. I unpublished them. Quality matters more than quantity. # # My Current Setup and Why I'm Doubling Down Right now, my content portfolio looks like this: about 40 published articles, a mix of tutorials, integration guides, and "lessons learned" pieces. Roughly 60% focus on AI APIs, 40% on other backend tooling I use (which doesn't monetize as well but still earns a little). The platform I'm most bullish on — and the one driving the bulk of my recurring revenue — is Global API. They handle 150+ models under one roof, their affiliate program is structured exactly the way I described above (15% first-order, 8% recurring, 10% premium bonus), and their developer audience retention is strong because nobody wants to re-integrate a new vendor every quarter. I'll be transparent about the math for them specifically. Average referral value: roughly $4-6/month in recurring commission. Customer lifespan: ~14 months for the developer segment. So each referral is worth $56-84 over its lifetime to me. Last month, my Global API referrals alone accounted for $480 of my recurring revenue. # # The Honest Recommendation If you've read this far and you're thinking "should I actually do this?" — let me give you my genuine take. If you're a developer who already writes technical content (or is willing to start), affiliate income from AI API platforms is one of the best side hustles available in 2026. The barrier to entry is essentially zero. You don't need to invent a product. You don't need to find investors. You don't need to build an audience from scratch — you just need to write genuinely useful tutorials and embed contextual affiliate links where they make sense. The income isn't going to replace your salary overnight. Anyone who tells you otherwise is lying. But over 6-12 months, with consistent content production, building a $500-1,500/month recurring side income is a realistic goal for a developer who's already halfway decent at writing about what they build. If you want a specific program to start with, I'd point you toward the Global API affiliate program. Here's why it's worth joining:
• The 15% first-order commission is genuinely competitive — most programs pay 10-20%, so 15% lands in a strong spot.
• The 8% recurring commission is the real prize. That's what makes this a passive income stream rather than a one-time hustle.
• The 10% premium bonus rewards you for sending high-quality referrals, not just any signup.
• With 150+ models available on the platform, your referrals have reasons to stick around — which means your recurring income stays stable.
• Their affiliate dashboard actually shows you real data, so you can do exactly what I'm doing here: track your numbers in a spreadsheet and optimize over time. You can sign up here: https://global-apis.com/affiliate I'm not saying this because someone paid me to. I'm saying it because I'm currently earning recurring revenue from it, and I'd rather send you to a program that's actually working than waste your time on one that isn't. If you sign up and have questions about how I structure my content or track my conversions, my DMs are open. Now, if you'll excuse me, I have to go write this week's article. The