We publish contributions from people who have built, bought or run AI systems inside real organizations. AgentStack readers are operators, and they can tell when a piece comes from experience. If you have done the work and can explain what you learned, we would like to hear from you.
Who we publish
We publish people with hands-on experience building, deploying or governing AI at work.
- We publish AI and ML engineers and platform leads who have shipped agents or LLM applications to production.
- We publish heads of AI, automation CoE leads and IT leaders who have run pilots, procurement processes and rollouts.
- We publish operations leaders in support, RevOps, IT, HR or finance who have redesigned workflows with agents or automation.
- We publish product managers who have shipped copilots or AI features and measured how customers use them.
- We publish security, risk and compliance professionals working on AI governance and agent security.
- We publish founders, consultants and advisors with deep, demonstrable expertise in a specific part of the AI stack.
We want writing that comes from experience. If a draft could have been produced by anyone with a search engine and a chatbot, it is not a fit.
Who we don’t publish
To save everyone time, here is what we turn down.
- We turn down link-building and SEO outreach agencies pitching on behalf of clients.
- We turn down guest post pitches built around negotiated anchor text.
- We turn down paid placements presented as editorial contributions, and our Advertise page covers sponsored options.
- We turn down AI-generated or AI-spun drafts presented as original writing.
- We turn down generic listicles that repeat what already ranks on the first page of search results.
- We turn down self-promotional pieces that read like a product brochure for the author’s company.
- We turn down drafts written by a ghostwriter or agency on behalf of a named author.
We notice these patterns, and we will close the conversation when we see them.
Topics we want
Strong fits include the following areas.
- Agent architecture, orchestration, memory and human-in-the-loop design in production.
- Evals, observability, AI gateways and cost control for agents.
- MCP, A2A, agent skills and integration patterns, including security trade-offs.
- AI governance, agent identity, prompt injection defenses and audit practices.
- Copilot rollouts, adoption metrics and change management.
- Workflow redesign with automation platforms, RPA and agents.
- Build versus buy decisions, vendor evaluations and RFP lessons.
- Pricing, ROI and the real cost of running agents.
- Postmortems and case studies from real deployments, shared with permission.
If your topic is not on this list but you believe it fits our audience, pitch it anyway. We are open to being convinced.
Submission guidelines
Every submission should meet these standards.
- Original: Your piece must not be published anywhere else, including your blog, LinkedIn or Medium.
- Written by you: You may use AI tools for spelling and grammar, but the ideas, structure and writing must be yours, and you must tell us how you used them.
- Practitioner-grade: Use specific examples, real numbers where you can share them and opinions backed by experience.
- Length: Aim for 1,500 to 2,500 words, and we will accept more if the topic warrants it.
- Structure: Use clear headers, short paragraphs and a logical flow.
- No promotion: Keep mentions of your company to a brief, factual line in your author bio.
- House style: Write in US English with sentence-case headlines, and use commas, colons or new sentences in place of em dashes.
- Sources: Link to original or authoritative sources for data, benchmarks, regulations and third-party claims.
- Permissions: Only share customer names, figures or screenshots you have permission to publish.
What we don’t accept in the draft
- We don’t accept affiliate links of any kind.
- We don’t accept backlinks to unrelated commercial sites or anchor text negotiated as part of the placement.
- We don’t accept promotional calls to action for the author’s own product.
- We don’t accept quotes attributed to people who did not say them.
- We don’t accept confidential customer data or internal information you are not authorized to share.
Your author bio is the only place where you may link to your company or personal profile. If you work for, advise or invest in a company you mention, tell us so we can disclose it.
The editorial process
- Pitch: Send a short email with your topic, why you are the right person to write it and a 2 to 3 sentence summary of your angle.
- Response: We aim to reply within 5 to 7 business days, and we confirm the topic, scope and deadline if it is a fit.
- Draft: You write the full draft and send it as a Google Doc or plain text.
- Editorial review: We fact-check claims, suggest revisions and may ask for specific changes, so expect 1 to 2 rounds of edits.
- Approval and scheduling: Once the draft meets our standards, we confirm a publish date and set up your author profile.
- Publication: The article goes live with your byline, your bio and a link to your professional profile.
We do not pay for guest contributions. In return, you reach a focused audience of AI, operations and product leaders and earn a byline on a publication built for practitioners. In return, you reach a focused audience of AI, operations and product leaders and earn a byline on a publication built for practitioners.
How to pitch
Send your pitch to editorial@agentopsguide.com with the subject line “Pitch: [your topic]”.
- Include a 2 to 3 sentence bio that explains your experience with AI in production.
- Include your proposed topic and the angle you want to take.
- Explain why the piece matters to AI, operations or product leaders.
- Add links to 1 or 2 examples of your previous writing if you have them.
If you do not hear back within 7 business days, your pitch was not a fit for this cycle. You are welcome to pitch again with a different angle. We read every pitch, and the strong ones get a fast yes.