Browse categories
Explore
Fiverr Pro
English
$
USD
I will fix your lovable replit or vibe coded ai app
Evgene Doronin
About this gig
Your Lovable, Bolt, Replit, Cursor, or Claude-built app may look finished while still hiding broken auth, exposed secrets, unsafe database access, unreliable APIs, missing error handling, or deployment risks.
I review AI-generated web apps as a production engineer, not as a prompt operator. The goal is a clear answer to three questions: what can fail, what matters first, and what is safe to fix inside a bounded milestone.
I can review:
- authentication and authorization
- API keys, environment variables, and data exposure
- backend endpoints and input validation
- database rules and tenant isolation
- Stripe webhooks and billing flows
- retries, logs, queues, and failure handling
- deployment configuration and smoke tests
My background includes 10+ years in production software and AI SaaS work across FastAPI, Laravel, Next.js, PostgreSQL, Stripe, Retell AI, and deployed mobile and web products. I also built AIMoat, an AI chatbot security scanner.
This gig is for an existing codebase. New features, redesigns, full rewrites, and formal penetration tests are separate scopes.
Message me with your repo, stack, live URL, and main concern before ordering.
Get to know Evgene Doronin
Evgene Doronin
- FromSpain
- Member sinceNov 2022
- Avg. response time1 hour
Languages
English, Russian, Spanish
Full-stack developer with 10+ years in production and 8 shipped projects across the App Store, Google Play, and web. I build mobile apps (native iOS/Android, Flutter), SaaS platforms (Next.js, Laravel, FastAPI, PostgreSQL), and AI products. Rare specialties: Voice AI with Retell AI (2 production SaaS products with AI call handling) and BLE/IoT with custom Bluetooth protocols for hardware devices. Message me before ordering so I can understand your project first.
My Portfolio
FAQ
What counts as a small app?
Up to 10 user-facing routes, one frontend, one backend, one database, and one deployment environment. Monorepos, multiple apps, complex multi-tenant systems, or more routes need a custom offer.
Which AI builders do you support?
Lovable, Bolt, Replit, Cursor, Claude Code, v0, and similar tools, provided the generated source code is available. I review the code and deployed behavior, not the brand of the tool.
Do you fix everything found in the audit?
No. Basic is report-only. Standard includes one agreed critical issue. Premium includes up to three agreed critical issues. Larger remediation becomes a separate custom offer after the findings are known.
Is this a penetration test or compliance certification?
No. This is an authorized production-readiness and defensive code review. It does not provide a compliance certificate, legal opinion, or guarantee that an application has no vulnerabilities.
Will you work directly on production?
Only when there is no safer option and we agree on a backup, access scope, and release window. I prefer a branch or staging environment and provide verification notes before deployment.
Can you continue after the rescue sprint?
Yes. Once the audit defines the real scope, I can send a custom offer for additional fixes, missing production layers, or ongoing engineering support.
