CMU TA built an AI-coded autograder for 1,200 sheets — and became its prompt injection
moarbugs · x · 2026-10-23
Shrey Tiwari, TA for CMU's undergrad software engineering course, built an autograder entirely with AI agents — writing zero code himself — that has graded 1,200 paper sheets from 89 students. Nearly all his time went into specifying what he actually wanted and how students behave. His meta-observation: asking "are you even reading this?" means a language model was the reader, and his own post is now handwritten prompt injection in the autograder's threat model. Core thesis: AI can write all the code but can't automate what nobody told it about — only the world reveals that, and deciding what to do remains the engineer's job.
More from coding & agent
- DeepStation meetup demos OpenAI's post-DevDay Codex tools live for developers — MicahBerkley · 2026-10-23
- Open-source semantic search makes GitHub issue search 2x faster, 8x more accurate — jamesmurdza · 2026-10-23
- HarnessRouter open-sources a unified API to run Codex, Claude Code and more as agent backends — tom_doerr · 2026-10-23
- React Scan author teases new tool for long-standing React perf pain — aidenybai · 2026-10-23
- Coinbase launches agent page letting ChatGPT, Claude and Grok trade crypto and stocks for you — MurrLincoln · 2026-10-23
- Malware found in GitHub DaisyUI-MCP repo: luajit.exe trojan steals credentials — AveryFreeman · 2026-10-23