Welcome back to "Shelved Projects", a series where I open up the archives to share systems that were technically functional and highly optimized, b...
For further actions, you may consider blocking this person and/or reporting abuse
One small thing on the SQLite setup, since you went to the trouble of turning on WAL. SetMaxOpenConns(1) quietly cancels most of what WAL buys you, because WAL's whole trick is letting readers keep going while a single writer works, and a pool of one forces every read to queue behind every write anyway. The common fix is two pools against the same file, a writer capped at one connection and a separate reader pool that can open many. Keep the busy_timeout as the safety net and your simulated swarm reads stop stalling on status writes. Also, that ending genuinely got me, the tool for escaping the wall being stuck behind the same wall is rough.
That's exactly, why SQLite is just for dev environment, postgresql is for production and there... With the 12 part series, I've noticed that NDA format just work better, so might actually rewrite it to use NDA instead of SQL in general just for the sake of maximum security and performance? The SQLite is just for the dev environment when I built it, because it was still on a mock MCP-Lite, so data derived was still mock data, the main reason I didnt update it to include MCP-Lite, is because since then, i built NMCP and NDA, which could lower overhead and make it far more efficient than I could ever imagine, so worth expanding on when I get the time.
Yip, that's the crux of it all, I cant make money off it, because any money made, cant be paid out. I guess there's a chance that RemoFirst would route the money for me, but relying on a partner to provide payment, without a dedicated payment platform seems to me like a problem waiting to happen...
The gut punch is the ending. The tool that breaks the geographic wall is locked behind the same geographic wall. That’s not a footnote, that’s the whole shape of a deadzone the exit is gated by the thing you’re trying to exit “Architecture is easy, logistics are hard” is the most honest line a builder can write, and almost nobody writes it. We obsess over WAL tuning and ship the elegant part, then a banking rail nobody warned us about decides whether the thing lives or sits on a shelf fully operational. That’s the most frustrating kind of done. For what it’s worth, 1-to-1 matching instead of spraying 200 resumes is the right instinct it respects the employer’s attention, which is the actual scarce resource. Don’t let this one stay shelved, man. The need didn’t go anywhere.
Exactly, the goal was for it to be a trusted resource by reputation. If an employer gets 1 good candidate from it, they'll come back for more, because working through an EOR, means it's not an employee, it's a B2B agreement on paper, the EOR is the actual employer. Which means no stupid laws like "this position must first be vetted against 10 locals, then 10 within the EU, before any foreign candidate may be considered, or we fine you and force you to fire them" which was a nice little fine-script that makes it smarter than hiring locally
That’s the part I didn’t have visibility into the EOR isn’t just a payment pipe, it’s a legal reclassification that routes around the “must vet 10 locals first” rule entirely. So the wall isn’t just HR bias at the resume stage, there’s a second wall baked into labor law that most people never even see because they never get far enough to hit it. The bank rail being the thing still blocking you from collecting your own cut is brutal irony on top of that.
Yip... That's why the second things calm down, I'll contact RemoFirst (my first option) for a partnership + API linkage, so their addendum can be added to the contract and so my point of profit can be connected. $20 a month is far more than I want from the system, but it's a start, that's why as I scale prices down, my cut shrinks, so it earns less money per client and dynamically relies more on bulk than individual pricing. $20 covers roughly 100 candidates per month in cloud compute (resume + searches), which is a good enough factor for me, the idea isnt to make money, it's sustainability and consistency.
@xulingfeng Just FYI, incase they're a scam, I have no affiliation at all with that Sam Tech guy. Had to make that clear, cuz it's a bit odd that he talked about hiring for a US company, when I'm about to make a post on exactly that 🫠 and now I caught you with the click-bait red flag 😁
🤣🤣🤣