Skip to main content

Write for Promptfoo

If you enjoy Promptfoo, want to help others learn it, and would like to build your reputation as a writer, this is for you.

What we publish​

We publish practical guides with working code for engineers developing, securing, and evaluating AI systems—from LLMs and agent frameworks to multi-agent deployments.

Types of content we value:

  • Engineering deep dives - Novel benchmarks, evaluation methodologies, red teaming frameworks, safety testing protocols
  • Production system guides - Real-world Promptfoo deployments at scale, monitoring, observability, and performance optimization
  • Research papers - Academic work using Promptfoo for evaluation, with reproducible experiments and datasets
  • Tutorials and guides - Step-by-step instructions with exact commands and reproducible examples
  • Video content - YouTube tutorials, conference talks, and technical walkthroughs
  • Integrations - How to use Promptfoo with AI frameworks (LangChain, LlamaIndex, Crew AI, OpenAI Agents SDK), cloud providers (OpenAI, Anthropic, Google Vertex), CI/CD pipelines, and security tools
  • Case studies - Real-world deployments, metrics, and lessons learned from production systems
  • Best practices - Test design, guardrails, data quality, and reproducibility
  • Troubleshooting - Clear solutions to common problems

Avoid broad AI explainers. Go deep, show code, and report results.

Topics we want now​

  • Production-scale Promptfoo deployments with monitoring and observability
  • Building red teaming workflows with Promptfoo in CI
  • Novel benchmark designs for multi-agent system evaluation
  • Evaluating model upgrades safely across providers (GPT-5, Claude 4, Gemini 2.5)
  • Reproducible jailbreak testing with shared datasets and academic rigor
  • Guardrail comparisons using Promptfoo evals
  • Promptfoo + OpenAI Agents SDK/CrewAI integration with live repo and CI
  • AI security testing workflows and best practices
  • Performance optimization for large-scale evaluation systems
  • Migrating from ad-hoc tests to evaluation plans

Compensation​

We pay for quality technical content. Deep engineering posts with working code get more than basic tutorials.

Writing guidelines​

  • Keep it clear and specific
  • Include runnable code with exact commands
  • Use version pins for models, SDKs, and Promptfoo
  • Add screenshots when helpful (include alt text)
  • Test all steps and provide a repo or example files
  • Follow our existing documentation style

Write in your own voice. AI-assist is fine, but we reject copy-paste ChatGPT output.

Maintenance: For 30 days after publication, we may request minor updates. Substantial revisions will be scoped separately.

How to pitch​

Email [email protected] with:

  • Topic and brief outline
  • Estimated length
  • Samples of your writing or relevant projects

You will receive a response anywhere from a few days to two weeks.

Process​

  1. Pitch review - We evaluate your proposal and provide feedback
  2. Draft submission - Submit as GitHub PR or Google Doc
  3. Editorial review - We review and provide feedback
  4. Final publication - After final checks, we publish on our site
  5. Promotion - We promote your work on social media, newsletter, and Discord

Rights and licensing​

By submitting content, you grant Promptfoo a non-exclusive license to publish and distribute your article. You retain full ownership and may republish your work elsewhere. We ask that you use our URL as the canonical source when cross-posting.

No affiliate links are permitted. Please disclose any conflicts of interest.

Need help?​

If you have questions about contributing or need help getting started, reach out to Tabs on Discord.

FAQ​

Q: May I cross-post to other websites?
A: Yes, but only on sites that support canonical URLs. Use our link as the canonical URL.