Show HN: 我开发了一个 AI 工具来评估我在 AngelList 上的交易流程
1 分•作者: stiline06•7 个月前
我是凯尔，一名软件工程师，通过 AngelList 联盟开始天使投资（大约 25 笔交易，每笔 1-10k 美元）。我看到过一些有趣的创意和聪明的创始人，但不太确定该关注什么，或者如何比较它们。我想要一个更系统地思考交易的系统。一个可以挑战我初步判断的第二意见。<p>它的功能：- 粘贴一份交易备忘录 → 获得 8 个标准（创始人、市场、发展势头等）的评分 - 每个评分都引用具体证据。“强劲的留存率”如果没有数据 = 较低的评分 - 并排比较交易，提出后续问题<p>演示：<p><a href="https://www.loom.com/share/a360b329f9e849c38c1ea70ba510d178" rel="nofollow">https://www.loom.com/share/a360b329f9e849c38c1ea70ba510d178</a><p>技术：- 使用 Claude Sonnet 4.5 进行分析（Anthropic 用于细致的判断） - 本地匿名化——在 API 调用之前，在客户端擦除公司/创始人的姓名 - 多层质量保证：准确性检查器捕捉幻觉，错误时自动重试，最终润色<p>我的经验：AI 编码工具让修修补补变得太容易了。我曾经同时进行 3 个修复，结果制造的错误比解决的还多。我不得不强迫自己慢下来，有条不紊地工作。更大的教训是：我花了几个月的时间独自调整，而不是获得外部反馈。这篇文章是我打破这种习惯的方式。<p>试用：免费层级每月有 20 次初步评估 + 3 次深度分析。我很乐意收到关于评分是否校准的反馈，也欢迎大家在这里讨论我开发的任何方面。<p><a href="https://angelcheck.ai" rel="nofollow">https://angelcheck.ai</a>
查看原文
I'm Kyle, a software engineer who started angel investing through AngelList syndicates (~25 deals, $1-10k each). I'd see interesting ideas and clever founders but wasn't sure what to look for or how to compare them. I wanted a system to think through deals more systematically. A second opinion to challenge my initial read.<p>What it does: - Paste a deal memo → get scoring on 8 criteria (founder, market, traction, etc.) - Every score cites specific evidence. "Strong retention" without numbers = lower score - Compare deals side-by-side, ask follow-up questions<p>Demo:<p><a href="https://www.loom.com/share/a360b329f9e849c38c1ea70ba510d178" rel="nofollow">https://www.loom.com/share/a360b329f9e849c38c1ea70ba510d178</a><p>Tech: - Claude Sonnet 4.5 for analysis (Anthropic for nuanced judgment) - Local anonymization—company/founder names scrubbed client-side before API calls - Multi-layer QA: accuracy checker catches hallucinations, auto-retry on errors, final polish<p>What I learned: AI coding tools make it too easy to tinker. I'd have 3 fixes going at once, creating more bugs than I solved. Had to force myself to slow down and work methodically. Bigger lesson: I spent months tweaking in isolation instead of getting external feedback. This post is me breaking that habit.<p>Try it: Free tier has 20 triages + 3 deep analyses/month. I'd love feedback on whether scoring feels calibrated and happy to talk about any elements of my development here.<p><a href="https://angelcheck.ai" rel="nofollow">https://angelcheck.ai</a>