reddit · SaaS · 10-05 05:50 38 分 ← 返回列表

I built a small prototype for evaluating whether engineers can actually do the work they claimed they can. looking for hiring managers and hiring related people to test it

hey everyone, i've been researching how engineering teams evaluate candidates, especially with resumes, take homes, and ai making traditional assessments harder to rely on. I built a small prototype around a different approach: Give the candidate a realistic engineering task → evaluate the implementation → change the requirement → see whether they can adapt without breaking existing behavior. The prototype currently captures things like: automated test results hidden validation

五维评分

付费意愿

需求真实度

个人可落地

竞争空间

新鲜度

AI 分析

痛点：AI 让传统简历、take-home 和笔试越来越难判断工程师真实能力，招聘方缺少能验证候选人能否适应需求变更的评估手段。

目标用户：工程经理、招聘经理、CTO、技术招聘人员等负责工程岗位招聘的人。

证据："I'm not looking to sell anything right now. I'm trying to find out whether this actually provides useful signal to people who hire engineers." 以及 "I'm especially interested in hearing what you currently use to determine whether someone can actually perform the engineering work they claim they can."

变现：按席位或按评估次数收费的 B2B SaaS（企业招聘预算），后期可加测评题库与报告服务。

MVP：做一个小型网页工具：预置 3-5 个真实工程任务，候选人提交代码后由 LLM 加自动化测试评分，然后自动改变需求并要求增量提交，输出对比报告给招聘方。两周内可完成单任务闭环与报告页面。

竞品：HackerRank、CodeSignal、CoderPad、Triplebyte

摘要：发帖人做了一个工程招聘评估原型：给候选人真实任务、评估实现、再变更需求看其能否在不破坏现有行为的前提下适应。目前处于免费找招聘方试用并收集反馈阶段，没有收入或明确付费意愿证据，属于早期验证中的想法，需求真实性尚未被他人附和验证。