Latest / The Side Hustle Career Podcast with Fexingo: Second Income, Freelancing While Employed / How Senior Engineers Monetize Technical Debt Cleanup
Transcript
- Lucas: We spend a lot of time talking about building things from scratch on the side, like launching apps or starting agencies. But there is a quieter, much more lucrative market for people who already know how to tear things apart professionally. I want to talk about technical debt cleanup. Luna: Right, because every engineering team has legacy code that nobody wants to touch, yet everyone needs fixed. Lucas: Exactly. And here is the specific case study: a senior backend engineer at a mid-sized fintech firm who started doing weekend code audits for smaller startups. He charges four hundred dollars an hour, does about ten hours a month on the side, and nets roughly forty thousand dollars a year in pure profit before taxes. Luna: Four hundred an hour is steep for a side gig, but if he is finding critical security flaws or performance bottlenecks, that price makes sense immediately. Lucas: It does, because he isn't just reading code. He is using automated static analysis tools to scan the repository, then he manually reviews the top twenty percent of flagged issues. He delivers a PDF report with actionable items, prioritized by severity. The client gets clarity without hiring a full-time consultant. Luna: So he is selling peace of mind and risk mitigation, not just lines of code changed. Lucas: Precisely. And the beauty of this model for someone employed full-time is that he doesn't write production code for the client. He doesn't take ownership of the repository. He provides the diagnosis. That distinction is crucial for avoiding conflicts of interest with his day job. Luna: That is smart. If he was fixing bugs, he might be stepping on toes or violating non-compete clauses. A diagnostic report is cleaner territory. Lucas: It is, but it requires a very specific skill set. You cannot just be a good coder. You have to be able to communicate technical risk to non-technical founders. His report includes an executive summary written in plain English, explaining why the database schema is fragile and what the business impact would be if it crashed during peak traffic. Luna: That translation layer is where the real value lives. Most engineers fail because they dump raw logs on a founder and expect them to understand. Lucas: He learned that through trial and error. His first three clients rejected him because his reports were too dense. He had to simplify the language, remove the jargon, and focus entirely on ROI. He frames every finding as a potential loss of revenue or a compliance risk. Luna: So he positions himself as a business advisor who happens to speak Python, rather than just a developer who writes tests. Lucas: Exactly. And this brings up the question of how he finds these clients. He didn't use Upwork or Fiverr. Those platforms are terrible for high-ticket consulting because you end up competing on price. He went straight to LinkedIn and targeted CTOs at seed-stage startups. Luna: Seed-stage founders are usually drowning in work and terrified of their own infrastructure collapsing. They don't have the budget for a big firm like Palantir, so they need a solo expert. Lucas: They do. And he posts content about common architectural mistakes he sees in early-stage companies. Not generic advice, but specific examples like 'why your microservices strategy is slowing down deployment.' It attracts the right audience organically. Luna: Content marketing for engineers is often ignored, but it works perfectly for niche expertise. You become the go-to person for a specific problem. Lucas: One thing I want to highlight is the tooling. He uses SonarQube for initial scanning, which is free for open source, but he pays for the enterprise version to get deeper historical data. Then he writes custom scripts in Python to parse those results into a standardized format. This automation means he can process a medium-sized codebase in about two hours instead of eight. Luna: Automation scales the diagnosis. So his margin improves as he gets better at configuring those tools, right? Lucas: It does. By month six, he had a template system. He could spin up a basic audit in under three hours total. That efficiency gain is what allows him to keep his hourly rate high while delivering the work quickly. He values his time, and the tools protect it. Luna: This feels like a sustainable model. It doesn't burn him out because he isn't on call for the client's systems. He does the work, hands over the report, and walks away. Lucas: He explicitly avoids on-call duties in his contract. If the client wants implementation support, he refers them to other freelancers and takes a small referral fee. He stays in the advisory lane. This keeps the scope tight and prevents scope creep, which kills most side hustles. Luna: Scope creep is the silent killer. Keeping the deliverable finite is essential for anyone trying to balance this with a full-time job. Lucas: Let's talk about the legal side again, because it bears repeating. He checks his employment contract carefully. Most standard agreements allow for outside work as long as it doesn't compete with the employer or use company resources. Since he is auditing third-party startups, not his employer's competitors, he is safe. But he never runs the scans on his employer's private servers using his personal account. Luna: Using separate hardware or cloud instances is non-negotiable. Mixing personal and professional environments creates liability nightmares. Lucas: Agreed. He uses a dedicated laptop for this work, wiped clean of any employer data. It is a physical boundary that protects both his sanity and his legal standing. It is a small detail, but it matters immensely. Luna: Physical separation is underrated. It stops the mental bleed-over between your day job and your side project. Lucas: Another interesting point is pricing structure. He offers two tiers. A basic scan for twenty-five hundred dollars, which includes the automated report and a thirty-minute call. And a deep dive for seven thousand five hundred dollars, which includes manual review of critical modules and a follow-up session after they implement changes. Most clients start with the basic tier. Luna: The upsell path is clear. Once they see the value of the basic report, they realize they need help implementing the fixes, which opens the door for the higher-tier engagement. Lucas: But he doesn't pressure them into the deep dive. He lets the findings speak for themselves. This builds trust. Founders are skeptical of consultants who push expensive packages. His restraint makes him feel like an ally rather than a salesperson. Luna: Trust is the currency in consulting. If you come across as hungry, you lose credibility. If you come across as helpful, you build a referral network. Lucas: He has actually started getting referrals from other freelancers now. Developers who don't want to do the audit work send clients to him, and he sends clients who need coding help back to his network. It is a symbiotic ecosystem. Luna: That is the holy grail of side hustles. You stop hunting for clients and start attracting them through reputation and peer recommendation. Lucas: Before we close this out, I want to mention that if we've helped you think through a career move or decision, consider supporting the show. Listener donations keep these episodes ad-free and allow us to dig into cases like this one. You can find us at buy me a coffee dot com slash fexingo. It really helps sustain the research behind these stories. Luna: Thanks for keeping the signal clear amidst all the noise. Now, back to the topic. Lucas: This model of technical debt cleanup is accessible to many senior engineers, but it requires a shift in mindset from builder to auditor. It asks you to value diagnosis as much as creation. Luna: And it proves that sometimes the most valuable thing you can offer isn't a new feature, but a clearer view of the existing reality. Lucas: If you are sitting on years of experience in a specific stack, ask yourself what problems you solve fastest. Maybe the answer isn't building something new, but diagnosing what is already broken.