Entry-Level AI-Assisted Developer (Apprentice Sprint) - Python/FastAPI

AI Developer Apprentice / AI-Leveraged Builder — Full Potential AI Full Potential AI is building practical systems where humans and AI agents work together to create measurable progress. We are expanding a small AI developer bench for people who can take a clearly defined ticket, use modern AI coding tools intelligently, build or repair a small capability, and return evidence that it works. This is apprentice-friendly, but strong experienced developers are welcome. We care more about useful shipped results, clear communication, and learning speed than impressive titles. What you may work on • Small Python/FastAPI services, scripts, integrations, and automation workflows. • AI-agent tools, MCP-style connectors, APIs, Telegram/WhatsApp, Notion, GitHub, Airtable, webhooks, and dashboards. • Reproducing bugs, testing fixes, documenting changes, and improving reusable capabilities. • Turning a vague problem into a small spec, implementation, test, and handoff. How the work operates 1. We issue one clearly bounded ticket at a time. 2. You confirm assumptions and estimated effort before coding. 3. You may use Codex, Claude Code, ChatGPT, Cursor, or equivalent, but you remain responsible for reviewing the output. 4. You return code, test evidence, limitations, and a short handoff note. 5. Work is retained when it moves a real goal forward and can be reused. Starting arrangement • Base rate: $5/hour for an initial paid 5-hour test. • Usually up to 5 hours/week at first; more responsibility can be earned through reliable results. • Future rate or scope changes are manual and require approval in advance. No automatic raises. • Manual time is disabled. Work must be tracked through the approved Upwork workflow. • No work begins without an assigned ticket and clear acceptance criteria. Strong fit • Can work independently from written instructions and ask precise questions when blocked. • Has built something with Python, FastAPI, JavaScript/TypeScript, APIs, automation, or AI tools. • Can read an existing codebase without immediately rewriting everything. • Understands Git, testing, debugging, and secure credential handling. • Is comfortable saying “I do not know yet” and investigating methodically. Please include 1. Your current hourly rate and lowest sustainable rate for a 5-hour test. 2. Two small examples of software, automation, API, or AI-agent work you personally completed. 3. Which AI coding tools you use and how you verify their output. 4. How you would investigate an AI agent that repeats a task but sometimes stops before completing it. 5. Your weekly availability and earliest start date. The first test will be small and practical. We are looking for someone who can help create a dependable bridge between human judgment, AI agents, and working software.