warkanlock

Hey HN,

We’re building Merlin, an assessment SDK for educators, that intent to help educators go from “materials” > “exam” > “results” without stitching together five different tools

What it does today:

1. Generate exams using AI (including from PDF/PPTX source material)

2. Share exams via links with specific users

3. Run/monitor exams in real time (focus on exam integrity / anti-cheating)

4. Evaluate submissions (auto-grading where possible) and track results

5. Analytics on outcomes so you can see what students actually struggled with

Check it out on: https://trymerlin.ai/

If you run assessments at any scale (university, bootcamps, training), I’d love to provide for free the license to use it in exchange for feedback on where existing workflows break: authoring, question quality, grading, proctoring, or reporting.

Most of the time we read about the worst bug we have faced; but this time I want to ask you something a little different: what was the worst known bug you introduced and why?

I have been trying to get (and compare) this information somewhere, but among technical blogs and the like I can't find much information where I can make a “sensible” comparison.

It would be good to know the size of your company and how much money you currently spend on infrastructure.

Yesterday I was watching Threads (1984) and an inherent question came to mind: why aren't we so scared anymore?