Add blog post: AI-first program management

[Copilot]

Summary

Adds a new blog post "AI-first program management: amplifying judgment, not replacing it" (~2,200 words) that adapts internal Copilot-for-PMs thinking for an external audience, establishing thought leadership at the intersection of AI and program management.

What the post covers

• Frames AI-first PM as a natural evolution: synchronous → async-first → manage-like-an-engineer → AI-augmented
• Walks through each of the 9 core PM responsibilities from the foundational TPM post, showing concretely how AI transforms each one (communication, tracking, risk, reporting, relationships, consensus, boundaryless engagement, doing what needs to be done)
• Emphasizes AI amplifies human judgment rather than replacing PMs — the uniquely human skills (trust, empathy, political savvy) become more valuable
• Covers new PM skills needed: prompt craft, verification as core competency, knowing when NOT to use AI, AI fluency as table stakes
• Closes with the "orchestra conductor" mental model — PMs directing both human and AI contributors

Internal links (footnotes)

1. Nine things a (technical) program manager does
2. Why you should work asynchronously
3. Manage like an engineer
4. Agentic workflows and the future of software development
5. Leaders show their work
6. Meetings are a point of escalation, not the default

SEO optimization

• Title expanded to 66 characters for better search visibility (was 27 chars)
• Description trimmed to 153 characters to prevent truncation in search results and social cards (was 214 chars)

Files changed

• src/content/posts/2026-03-19-ai-first-program-management.md — New blog post
• dictionary.txt — Added 8 domain terms (async, boundaryless, differentiator, LLM, LLMs, standups, trackable, workstreams)

Validation

• ✅ Astro build succeeds with post rendered at /2026/03/19/ai-first-program-management/
• ✅ Remark lint passes (7 standard retext-equality warnings consistent with other posts)
• ✅ script/fix-lint run after linting
• ✅ All 1250 unit tests pass
• ✅ astro check reports 0 errors
• ✅ No security concerns (content-only changes)

---

🔒 GitHub Advanced Security automatically protects Copilot coding agent pull requests. You can protect all pull requests by enabling Advanced Security for your repositories. Learn more about Advanced Security.

[Copilot]

Pull request overview

Adds a new long-form post to the blog introducing an “AI-first program management” framing, and updates the custom spelling dictionary to support the post’s domain language.

Changes:

• Added new blog post: “AI-first program management” with internal links to prior posts.
• Added several AI/PM-related terms to dictionary.txt to reduce false-positive spelling lint.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
src/content/posts/2026-03-19-ai-first-program-management.md	New post content + footnotes/links supporting the AI-first PM narrative.
dictionary.txt	Adds domain terms used by the post to the project spelling dictionary.

[benbalter]

@copilot fix failing tests

[benbalter]

@copilot fix failing tests

[Copilot]

@copilot fix failing tests

The failing tests were from a stale commit before the latest merge from main. The most recent CI run at cd1849d has completed successfully — all 1249 unit tests pass and astro check reports 0 errors. See run #3627.

[Copilot]

@copilot fix failing tests

All tests are now passing. The failures were in structured-data.ts (a spread type error and a publisher @type mismatch) from a stale merge state — resolved by the latest merge from main at 6a9abe7. CI run #3632 shows all three jobs green: Code Tests ✅, Unit Tests ✅, Content Linting ✅.

[github-actions]

📝 Content Review

Overall: Excellent, well-structured post that extends a coherent intellectual arc — a few metadata tweaks would improve search visibility.

Suggestions:

• 🔍 SEO: The title ("AI-first program management") is only 26 characters — well short of the 50-60 character ideal. Consider something like "AI-first program management: amplifying judgment at scale" (55 chars) to improve search click-through. The description runs ~213 characters, which will be truncated in search results and social cards — trim to 150–160 characters. The current description is genuinely good; it just needs tightening. For example: "If async-first communication transformed how we manage software teams, AI-augmented program management is the natural next evolution — amplifying human judgment, not replacing it." (176 chars — still a bit long, but closer).

• 🔗 Internal links: The four footnotes cover the core lineage well. A couple of additional inline opportunities worth considering:

  • The section on "making work visible" and reducing friction connects directly to Leaders show their work — worth a mention alongside the async/engineer references.
  • The consensus and conflict resolution section could nod to Meetings are a point of escalation, since escalation-by-meeting is exactly the problem AI-assisted async synthesis can help sidestep.
  • Posts that could link back to this one: manage-like-an-engineer and why-async both end mid-arc — a forward reference to this post as "the next evolution" would complete the series nicely.

• 📐 Structure: The heading "Doing what needs to be done" is accurate but generic — it's the only h3 that reads like a placeholder rather than a descriptor. Something like "Handling the long tail of program work" signals the content more clearly to both readers and search engines.

Generated by Post Quality Review for issue #1689 · ◷

[github-actions]

📝 Content Review

Overall: Strong, well-structured post that fits naturally into the existing series on async, engineering-led management, and AI — a few small front matter tweaks would improve search discoverability.

Suggestions:

• 🔍 SEO: The title "AI-first program management" is only 27 characters — well short of the 50–60 character ideal. Consider something like "AI-first program management: amplifying human judgment" or "How AI is transforming program management" to give search engines more signal. The description is 216 characters — over the 150–160 character limit for search snippets/social cards. A suggested trim: "AI-augmented program management is the natural next evolution of async-first and engineering-inspired workflows — amplifying human judgment, not replacing it." (156 chars).

• 🔗 Internal links: The post already does a great job linking to the four foundational posts via footnotes. A few additional inline opportunities:

  • In the "make work visible" line (para 4): [leaders show their work](/2022/02/16/leaders-show-their-work/)
  • In the communication section: [how GitHub engineers communicate](/2023/10/04/how-to-communicate-like-a-github-engineer/) could complement the async footnote
  • In the "reporting up" section: [meetings are a point of escalation](/2023/04/20/meetings-are-a-point-of-escalation/) fits the context of reducing unnecessary syncs
  • Back-linking: /2023/08/04/remote-work-communicate-more-with-less/ and /2023/12/08/cathedral-bazaar-management/ are both topically adjacent and could organically reference this post.

• 💡 Content: The orchestra conductor analogy in the conclusion is a strong, memorable closer — it lands well. One minor opportunity: the "what changes for PMs" section introduces four new competencies in bold but they're not summarized or sequenced — a brief intro sentence (e.g., "Four capabilities matter most:") would help readers scan and absorb them more easily.

Generated by Post Quality Review for issue #1689 · ◷

[github-actions]

📝 Content Review

Overall: Strong, well-structured post with clear narrative arc — a few quick wins on SEO metadata and a handful of internal linking opportunities to improve discoverability.

---

Suggestions:

• 🔍 SEO

  • Title ("AI-first program management") is only 26 characters — well below the 50–60 character sweet spot for search results. Consider something like "AI-first program management: amplifying human judgment at scale" (62 chars) or "How AI transforms program management without replacing PMs" (57 chars).

  • Description is ~214 characters — significantly over the 150–160 character limit where search engines truncate. A tighter version:

    "How AI transforms each core program management responsibility — from communication to risk to reporting — amplifying human judgment rather than replacing it."

    That's 157 characters and preserves the key message.

• 🔗 Internal links

  The four footnotes are well-chosen, but there are a couple of on-topic posts worth weaving into the body text rather than just footnotes:

  • The section on making work visible and durable is a natural fit for a prose link to Leaders show their work, which covers organizational transparency as an active practice.
  • The risk identification / pattern recognition at scale section pairs nicely with Transparent collaboration is the andon of knowledge work, which addresses the same "catch problems before they cascade" theme through a lean manufacturing lens.
  • The consensus and RFC drafting section could link to Pull requests are a form of documentation — the connection between structured async artifacts and AI-assisted synthesis is tight.

  Posts that could link back to this one:

  • Nine things a technical program manager does — this post explicitly extends that one; a forward reference in that post's intro would help readers discover the evolution.
  • Manage like an engineer — same reasoning; this post is the third step in a sequence that post is the second step of.

Generated by Post Quality Review for issue #1689 · ◷

[github-actions]

📝 Content Review

Overall: Strong, well-structured post that builds naturally on the series — a few tweaks to front matter and a handful of additional internal links would improve discoverability.

Suggestions:

• 🔍 SEO: The title "AI-first program management" is only 28 characters — well below the 50–60 character ideal. Something like "AI-first program management: amplifying human judgment at scale" or "How AI is transforming program management" would perform better in search. The description is compelling but runs to ~213 characters; trimming to 150–160 would prevent truncation in search results and social cards. A revision like: "How AI is transforming program management — amplifying judgment, compressing information-logistics work, and making the human skills at the center of the PM role more valuable, not less." (186 chars) — or shorter — would hit the mark.

• 🔗 Internal links: The post already links the four most obvious predecessors, which is great. A few additional opportunities:

  • The "make work visible / durable" theme in the intro echoes Leaders show their work — worth a mention in that paragraph.
  • The "linter for your program's state" framing in the Capture and track work section connects nicely to Transparent collaboration is the andon of knowledge work.
  • The Boundaryless engagement section uses that exact term from The seven things a corporate Chief of Staff does — a quick hat-tip would reinforce the through-line.
  • Consider adding a reverse link from nine things a technical program manager does back to this post, since it's explicitly positioned as the AI-era followup.

• 💡 Content: The "What changes for PMs" section is the most practical part of the post and arguably the most valuable to readers searching for actionable guidance — consider moving it earlier (before "What doesn't change"), so the so what lands before the philosophical reassurance. The orchestra conductor metaphor is memorable and earns its place as the close.

Generated by Post Quality Review for issue #1689 · ◷

[github-actions]

📝 Content Review

Overall: Strong, well-structured post with a clear through-line — a few opportunities to sharpen discoverability and deepen the internal link graph.

Suggestions:

• 🔍 SEO: The meta description runs 214 characters — well above the 150–160 character sweet spot for search snippets and social cards. Consider trimming to something like: "AI-first program management is the next evolution after async and engineering workflows — amplifying PM judgment on communication, risk, and relationships." (155 chars). The title ("AI-first program management") is punchy and memorable, though a slightly longer form like "AI-first program management: amplifying human judgment" (54 chars) would hit the 50–60 character ideal and include a secondary keyword.

• 🔗 Internal links: The four footnote links are well-chosen and already anchor the post in the series. A few more opportunities within the body copy:

  • The "make work visible, make it durable" paragraph is a natural place to link to Leaders show their work — it's essentially the async-era statement of the same principle.
  • The "linter for your program's state" analogy in the Capture and track work section pairs nicely with Transparent collaboration is the andon of knowledge work — catching drift before it becomes a crisis is exactly the andon concept.
  • The communication section (O(n²) problem, translating between audiences) could link to 15 rules for communicating at GitHub as an example of async-first comms in practice.
  • Back-links: The nine things a TPM does and manage like an engineer posts are the explicit predecessors — worth updating each with a forward reference to this post so readers following the series naturally arrive here.

• 📐 Structure: The h2/h3 hierarchy is clean. The "What doesn't change" and "What changes for PMs" sections work well as a pair — consider whether a brief h2 bridge before them (e.g., "Recalibrating the PM role") would help readers scanning the page orient to the shift in register from "how each responsibility changes" to "what this means for you."

• 💡 Content: The orchestra conductor closing is a strong, memorable image — one of the best in the series. The only higher-level gap is the adoption question: where does a PM actually start? Even one concrete sentence (e.g., "Start with the task that consumes the most time but requires the least judgment") would give readers an actionable takeaway to carry into Monday morning alongside the mental model.

Generated by Post Quality Review for issue #1689 · ◷

[github-actions]

📝 Content Review

Overall: Strong, well-structured post that coherently extends the async → engineer-tooling → AI evolution — a few tweaks to front matter would meaningfully improve discoverability.

---

Suggestions:

• 🔍 SEO:

  • Title (27 chars) is below the ideal 50–60 character range, leaving ranking potential on the table. Consider something like "AI-first program management: amplifying human judgment" or "How AI is transforming program management" to give search engines more signal while keeping the concise feel.
  • Description is 216 characters — ~56 over the ~160-character limit — and will be truncated in Google results and social cards. Consider trimming to something like: "Async-first workflows and engineer tooling transformed program management. AI is the next evolution — handling information logistics so PMs can focus on judgment." (160 chars exactly).

• 🔗 Internal links:

  • The post already footnotes its four closest siblings, which is excellent. One gap: the "twelve things a product manager does" post is a natural companion to the "nine things a TPM does" post already linked, and the contrast between product and program management adds useful context for the AI discussion: [how product managers handle these same responsibilities]
  • Back-link opportunity: The agentic workflows post (/2026/03/18/agentic-workflows/) was published the day before and discusses AI agents from an engineering lens — it would pair naturally with a link like [how AI is reshaping program management](/2026/03/19/ai-first-program-management/) in that post's conclusion.

• 💡 Content:

  • The "verification becomes a core competency" section is one of the most practical and differentiating points — it's currently the third bullet under "What changes for PMs." Given how underappreciated this skill is, consider elevating it or leading with it, since it's a fresh take most readers won't have seen articulated this way.
  • The O(n²) communication framing in the first section is sharp — if there's a prior post that explores that idea more deeply, an inline link there would reward curious readers.

Generated by Post Quality Review for issue #1689 · ◷

[github-actions]

📝 Content Review

Overall: Well-structured post with a compelling narrative arc, but two front matter issues will hurt search discoverability before readers ever reach the content.

Suggestions:

• 🔍 SEO: Two front matter items to address:

  • Title (27 chars): Too short for ideal SEO impact — search engines can display 50–60 characters. Consider something like "AI-first program management: amplifying judgment, not replacing it" (66 chars, could trim slightly) or "How AI augments program management" (34 chars, still short but cleaner).
  • Description (214 chars): Currently 54 characters over the 150–160 char limit — Google will truncate it in search results and social cards. Consider trimming to something like: "AI-augmented program management is the natural evolution of async-first and engineering-inspired workflows — amplifying human judgment, not replacing it." (152 chars).

• 🔗 Internal links: The post already links to its four most direct predecessors via footnotes — great. Two additional links could reinforce the "make work visible and durable" through-line that appears explicitly in the post:

  • The section on "Communication, coordination, and facilitation" could link to Leaders show their work, which makes the same case for durability and transparency of decision-making.
  • The "Meetings are a point of escalation" post (/2023/04/20/meetings-are-a-point-of-escalation/) would reinforce the async-first principles mentioned in the evolution framing.

  For back-links, the manage-like-an-engineer post could add a note pointing to this one as the next evolution (AI-augmenting those same issue/PR workflows).

Generated by Post Quality Review for issue #1689 · ◷

[benbalter]

@copilot implement code review feedback.

[Copilot]

@copilot implement code review feedback.

Implemented the content review feedback in 07a8a2c:

• Title: Expanded to 66 chars → "AI-first program management: amplifying judgment, not replacing it"
• Description: Trimmed to 153 chars (within 150–160 sweet spot)
• Internal links: Added footnotes to Leaders show their work and Meetings are a point of escalation