1d ago · 3 min read · Bank statement parsing is one of the hardest document extraction challenges. Every bank uses a different format, different column layouts, and different terminology. Here is how fintech companies solve it. Why Bank Statements Are Hard Unlike invoices...
Join discussion1d ago · 3 min read · The default retry schedule for most payment processors is suboptimal. Here is what the data shows about when and how to retry failed charges for maximum recovery. Why Default Retries Fail Stripe, Paddle, and Recurly all have built-in retry logic. But...
Join discussion2d ago · 4 min read · Last year I set out to build a cafe discovery platform. The dataset I ended up with was 440,000 cafes across 195 countries. The first question was obvious: how do you rank them? Star ratings are broken. A 4.2 on Google Maps tells you almost nothing —...
Join discussion2d ago · 3 min read · I spent the last year building a knowledge graph that now holds 1.8 million entities and nearly 3 million connections between them. Not for a research lab or a big tech company — for an independent media site called DropThe. Here's what I learned abo...
Join discussion2d ago · 14 min read · The first two posts covered: CLI vs MCP is a fight over pipes — what's actually missing is the faucet (post one); Agent adoption is blocked by two layers of barriers — platform lock-in and organizational control (post two). This post looks at how dif...
Join discussion2d ago · 10 min read · In the previous post we argued that the CLI vs. MCP debate is really a debate about pipes — and what's actually missing is the faucet. This post digs one level deeper: even if you open the faucet, you probably still can't connect to it. The reasons a...
Join discussion
