Built from work this engineer published themselves. It makes no claim that they are available for hire, and it is marked noindex until they claim it. Source: simonwillison.net. This is you, or it should not be here? Write to hello@aiengineer.me.
ai-engineer
Simon Willison
Datasette and the llm CLI, plus a public archive of prompt-injection notes.
Co-created Django and now builds open-source tooling for working with language models: Datasette for exploring data and the llm command-line client for running prompts, embeddings, and plugins against local and hosted models. Publishes detailed running notes on prompt injection and on the exfiltration risks of tool-using assistants.
Technical stack & depth
Prompting
Prompt design, iteration, and instruction-following for a specific job.
Guardrails
Policy, refusal, red-teaming, and the cases the model must not invent.
Production hardening
Latency, cost, observability, and the path from a notebook to a service.
Engineered solutions
- Repositorysimonw/llmgithub.com
Author and maintainer of the llm CLI for running prompts, embeddings, and plugins against local and hosted models.
- Write-upPrompt injection write-upssimonwillison.net
Ongoing public catalogue of prompt injection attacks, exfiltration paths, and proposed mitigations.
Something here out of date or wrong? Write to hello@aiengineer.me and it will be corrected or removed.