100x Coding with Claude
Up until recently I’ve been using LLMs like copilot as a developer companion and I have mostly been disappointed. Maybe Claude can change this?
Update: This post received a lot of quality comments. Thank you! Based on your comments I’ve written a separate post offering solutions to the shortcomings mentioned in this post. You can find it here. Enjoy!
Bounties might increase work efficiency by sourcing specific talent for specific tasks. The fact that the bounty is paid on a per project basis reduces the probability of disagreement between the creator and solver.
O ne of the challenges is that some bounties might be completely greenfield projects. This means that if you want to solve it, you need to start from a scratch project. Which includes a loooooot of boilerplate, so much boilerplate that I’ve ignored bounties for a while because it takes so much energy just to get to the teal solving part.
I’ve heard a lot of great things about Anthropic’s Claude and especially Sonnet 3.5. So when a bounty for building something cool on top of the The Grid API presented itself I had an opportunity. An opportunity to use a state of the art LLM to discover how to only focus on the hard development challenges and hopefully…
