Live Laravel AI Support Agents
Budget / SalaryHourly project
TypeFreelance project
LocationRemote
Posted1 hour ago
I’m expanding an existing Laravel PHP module and need a couple of agentic AI routines that can handle automated customer support right inside the app. All coding must happen with me on a live screen-share—Zoom or VS Code Live Share—so you’ll be pair-programming in real time; no offline work or after-hours commits.
During each session we’ll design an agent, wire it to my current REST models, hit the staging database, and iterate until tests pass. The stack is Laravel 10, PHP 8.2, Livewire, Inertia, Git, and an OpenAI endpoint; familiarity with LangChain-style workflow orchestration in PHP is a plus.
Please reply with concise evidence of your hands-on experience building conversational or retrieval-augmented agents, especially inside Laravel or comparable frameworks.
Deliverables created live with me
• Two well-commented PHP classes that register as “agents” in my existing module and answer customer queries.
• Unit and feature tests demonstrating correct responses on a provided question set.
• Git commits pushed to my private repo after each green test run.
Acceptance criteria
• All code produced while we are connected.
• Agents run error-free under PHP 8.2 and pass the agreed tests.
• No unapproved proprietary dependencies.
If you can start immediately and enjoy fast, collaborative sessions, let’s book the first two-hour block.
During each session we’ll design an agent, wire it to my current REST models, hit the staging database, and iterate until tests pass. The stack is Laravel 10, PHP 8.2, Livewire, Inertia, Git, and an OpenAI endpoint; familiarity with LangChain-style workflow orchestration in PHP is a plus.
Please reply with concise evidence of your hands-on experience building conversational or retrieval-augmented agents, especially inside Laravel or comparable frameworks.
Deliverables created live with me
• Two well-commented PHP classes that register as “agents” in my existing module and answer customer queries.
• Unit and feature tests demonstrating correct responses on a provided question set.
• Git commits pushed to my private repo after each green test run.
Acceptance criteria
• All code produced while we are connected.
• Agents run error-free under PHP 8.2 and pass the agreed tests.
• No unapproved proprietary dependencies.
If you can start immediately and enjoy fast, collaborative sessions, let’s book the first two-hour block.
Apply on Freelancer →
Project sourced from Freelancer.com. Applications happen directly on the original platform — we never collect your data.