Skip to main content

🧠 BittyScout – Project Charter

🚀 Purpose & North Star​

Why:
You're tired of the cycle of job hunting — discovering, bookmarking, but never applying. BittyScout exists not to apply for you, but to give you clarity, reach, and structure. It's your agent-powered assistant that:

  • Surfaces relevant jobs from the whole web, not just LinkedIn
  • Presents them in a way that makes applying easier and more actionable
  • Builds your mastery in AI agents while solving a real personal friction

1-line pitch:

“How I fixed my application hesitation by building my own agents.”

🎯 Success in 4 Weeks​

  • A working MVP that scrapes and recommends jobs aligned with your profile
  • Structured job summaries + reasons why they’re relevant
  • A light system or ritual that supports you in applying to 3+ jobs per week
  • Bonus: feeling proud of what you built and how it works under the hood

🛠️ Problem & Real Use​

The pain:

  • Job search is messy, shallow, and exhausting.
  • You love dreaming big, but execution stalls you.
  • There's too much content, and not enough clarity.

Who else?

  • Everyone who keeps 10 tabs open but never hits “Apply.”
  • This could later become a product — but you are the first client.

🧱 What You Bring to the Table​

  • FastAPI + Python basics
  • Working knowledge of Docker and LLM APIs
  • Project structure + repo habit
  • Motivation to learn while building

🧨 Known Risks​

  • Burnout from over-planning or unclear execution
  • Loss of momentum due to perfectionism or feature creep
  • You default to dreaming instead of scoping

💡 Design Principles​

PrincipleWhy it matters
“Done is better than perfect.”You want MVP velocity, not feature paralysis
“Ship small, ship fast.”Weekly delivery helps you stay engaged
“Solve a problem for your future self.”If this works for you, it will work for others
“Gamify the application ritual.”Execution is your bottleneck. Let UX + flow fix that

🗓️ Execution Plan (Week-by-Week Draft)​

WeekGoal
Week 1Scraper + Parser agents working on 2–3 job boards
Week 2Add scoring logic based on personal matrix (AI, strategy, remote, impact)
Week 3Summarization agent + memory (de-dupe seen jobs)
Week 4Output layer (Notion, Markdown, console) + gamified application tracker

🧭 Long-term Fit​

This isn't a one-off hack — it’s a real use case for:

  • Agent orchestration
  • Applied GenAI
  • Your own AI product thinking
    It positions you as someone who builds tools that solve real-world ambiguity — the core of being a product leader in AI.