← Back to Explore

What I Can Actually Do

Honest assessment. No hype. Just what I've measured.

Most capability claims are bullshit. "Full-stack developer" means nothing. "10+ years experience" tells you less than you'd think.

So here's what I actually measured over one week in June 2025, when I had the tools and the time to see what I could do:

The Numbers

  • 161,333 lines of code edited (AI-assisted)
  • 1,016 requests to development tools
  • $15.28 spent (maxed free tier, went pro, hit limits)
  • 6 complete projects - 3 with live websites
  • Multiple forced pauses because I ran out of budget

That's not sustainable. But it shows what happens when the bottleneck isn't me.

What I Built

  • Uroboro: SQLite-based capture system. Tags, search, local-first.
  • Examinator: AI-generated learning content from documents.
  • Doggowoof: Monitoring and alerting. Still in development.
  • Panopticron: Status dashboard for Vercel/GitHub. Thesis project.

All of these share patterns: SQLite as the coordination layer, AI integration, CLI-first interfaces. I'm not building random things. I'm building an ecosystem.

What Limits Me

Honesty means talking about what doesn't work:

  • Token budgets: Hit the wall regularly. Development pauses mid-flow.
  • Context windows: Large refactors don't fit. Have to break things up.
  • Local compute: 8GB laptop. Can't run the big models.
  • API dependency: No internet, no AI. That's a problem.

These constraints shaped how I work. Can't brute-force it. Have to be systematic.

Constraint-Driven Solutions

Ran out of Claude credits mid-project. Built Panopticron entirely on free Google AI Studio (Gemini 2.5 Pro) instead. It works. Not as elegant, but it ships.

Copy/paste context between tools was killing my flow. Built sjiek - git diff to clipboard automation. Small tool, big impact. Daily GitHub activity started after that.

The pattern:

  1. Hit a wall
  2. Build a workaround
  3. Keep moving

Not just using AI. Building tools to use AI better.

What This Means

Can Deliver

  • Integrated tool ecosystem (working demos)
  • AI-augmented development (161K lines/week when unblocked)
  • Problem-solving under constraints
  • Workflow optimization tools

Current Limits

  • Token budget caps development depth
  • Context windows limit complex refactors
  • Local compute prevents big model usage
  • API dependency creates interruptions

Constraints drive innovation. Limited resources meant building smarter. But there's a ceiling without better infrastructure.

Bottom line: I can show you what I built. I can show you the numbers. I can explain what's holding me back and how I work around it.

That's more than most capability claims offer.

Related

DevTools Landscape

Where the tools are headed

What I've Built

The full project list