# We Had 13 Engineers Spend Three Months Finding Vulnerabilities with LLMs

DevFeed: [We Had 13 Engineers Spend Three Months Finding Vulnerabilities with LLMs](<https://devfeed.tech/articles/we-had-13-engineers-spend-three-months-finding-vulnerabilities-with-llms-29646.md>)

Original publisher: [Read original article](<https://goteleport.com/blog/finding-vulnerabilities-with-llms/>)

Author: info@goteleport.com (Rob Picard)

Published: 2026-08-19T00:00:00Z

Content type: article

Language: en

Sources: [Teleport](<https://devfeed.tech/sources/teleport.md>)

Topics: [Vulnerabilities](<https://devfeed.tech/topics/vulnerabilities.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [llms](<https://devfeed.tech/tags/llms.md>), [security](<https://devfeed.tech/tags/security.md>), [software](<https://devfeed.tech/tags/software.md>), [vulnerabilities](<https://devfeed.tech/tags/vulnerabilities.md>)

## AI overview

Teleport describes how 13 software engineers spent a quarter using frontier LLMs and coding agents to review an existing codebase, find vulnerabilities, triage findings, and fix bugs. The article reports that simple prompting of an LLM at one codebase component performed better than a complicated multi-stage agentic harness in their experiments.

## Source excerpt

Our approach to "pressure washing" our codebase using frontier LLMs.