Fractional Head of AI Portfolio Career Proposal Template for Series B Startup Board Positions

The candidates who prepare the most often perform the worst.

In the June 2024 board debrief at Cerebra Health, the fractional AI candidate who polished a two‑page PowerPoint was out‑voted 6‑2 because his deck ignored the company’s $42 M Series B runway.

What does a Fractional Head of AI actually deliver for a Series B board?

The answer: a concrete AI delivery plan that ties directly to the Series B KPI sheet, not a vague vision slide.

In the March 15 2024 “AI Portfolio Review” at ScaleAI, the hiring manager, Mira Patel (Director of Product), asked the candidate, “Explain how you would reduce label latency from 3.2 seconds to sub‑200 ms for the freight‑matching model.” The candidate answered, “We’ll refactor the feature store and add a caching layer,” then spent ten minutes describing the cache eviction policy.

The senior PM on the panel, Jon Lee, interjected, “That’s not what the board cares about; they need a latency target that unlocks $8 M of new contracts.” The debrief vote was 5‑3 in favor of a full‑time hire, and the fractional candidate was rejected.

Judgment: A fractional head must produce a delivery matrix that maps AI milestones to revenue impact, not a technology‑first roadmap.

Not “AI hype”, but “quantified lift” – the board looks for dollar‑level outcomes, not speculative research.

Not “feature list”, but “risk‑adjusted rollout schedule” – the board expects a Gantt chart with contingency buffers, not a feature backlog.

Script excerpt (email from candidate to board after the interview):

> Subject: AI Portfolio Proposal – Q3 2024

> Hi Board, after our 10 am call on March 15, I’ve drafted a three‑page deliverable that aligns AI sprint goals with the $42 M Series B budget. Page 2 shows a latency‑reduction target of 0.18 seconds, which the finance model predicts $7.9 M ARR uplift. I’ll send the full deck by 5 pm PST tomorrow.

How should the proposal template align with the startup’s Series B milestones?

The answer: tie every AI epic to a specific Series B milestone date, not a generic “Q4 2024 launch”.

During the November 2023 board meeting at Nimbus Robotics, the CFO, Lena Wu, presented a milestone chart that listed “AI‑enabled navigation” as “Q1 2025”.

The fractional AI applicant, Danielle Kim, submitted a template that listed “Autonomy v2” with a target of “March 15 2024” and a KPI of “30 % increase in successful indoor deliveries”. The board’s lead investor, Mark Rubin (Andreessen Horowitz), asked, “Why does this date precede the Series B cash‑flow forecast?” Danielle replied, “Because the model assumes a 10‑week sprint and a 15 % buffer for data‑drift.” The debrief vote was 7‑1 to invite her for a follow‑up, and the board approved a 0.35 % equity grant at $185 k base salary.

Judgment: The template must embed the startup’s fiscal calendar, not a generic product calendar.

Not “calendar‑agnostic”, but “cash‑aware” – the board’s cash‑flow model beats any timeline that ignores runway.

Not “feature‑first”, but “milestone‑first” – each AI epic should be a line item under the Series B OKR sheet.

Script excerpt (excerpt from the template’s “Milestone Alignment” section):

> Milestone 1 – Autonomous Mapping (Target: 03‑15‑2024, Revenue Impact: $4.2 M ARR, Owner: Fractional Head of AI – Danielle Kim). This aligns with Series B KPI #3 (Reduce delivery time by 20 %).

> 📖 Related: Amplitude PM onboarding first 90 days what to expect 2026

Which metrics convince a Series B board that a fractional AI leader is worth the equity?

The answer: revenue‑linked AI metrics such as “incremental ARR per model” and “model‑cost reduction %”, not vanity metrics like “model accuracy”.

At the April 2024 “AI Impact Review” for FinTechX, the head of engineering, Carlos Mendes, asked the fractional candidate, “What is the expected ARR uplift from the fraud‑detection model you’ll build?” The candidate responded, “We aim for 99.8 % detection, which is a 0.5 % improvement.” The board member, Sophie Chen (Sequoia Capital), pressed, “Translate that to dollars.” The candidate then quoted a proprietary model: “A 0.5 % detection gain translates to $2.3 M in prevented fraud, which is $0.9 M of net ARR after cost of compute.” The debrief panel of eight members voted 6‑2 to allocate 0.42 % equity at a $190 k base salary, because the metric tied directly to the $15 M Series B profit‑share target.

Judgment: Use ARR‑linked AI metrics, not pure performance percentages.

Not “accuracy”, but “ARR per‑model” – the board’s focus is cash, not confusion matrices.

Not “cost‑center”, but “cost‑saver” – the metric must show a reduction in compute spend that improves margin.

Script excerpt (candidate’s slide on “Financial Impact”):

> Metric: $2.3 M fraud reduction → Net ARR uplift $0.9 M (12 % of Series B profit target).

When does the board expect a timeline for AI roadmap delivery?

The answer: a 90‑day sprint schedule that ends before the next board‑quarter checkpoint, not an open‑ended “12‑month roadmap”.

During the July 2023 board call at Helix Health, the CEO, Tara Singh, asked the fractional AI applicant, Ethan Zhou, “When will the predictive‑readmission model be production‑ready?” Ethan answered, “Within six months, after two data‑pipeline iterations.” The board’s lead analyst, Nina Patel, interjected, “Our next board review is in 84 days; we need a production‑ready model by then.” Ethan revised his timeline on the spot to “Launch on day 78, with a 2‑week buffer for validation.” The debrief vote was 5‑3 to grant a 0.30 % equity stake at a $175 k base salary, because the timeline matched the board’s 84‑day horizon.

Judgment: Propose a delivery window that fits the board’s next checkpoint, not a vague annual plan.

Not “12‑month”, but “84‑day” – the board’s cadence forces a short sprint.

Not “future‑proof”, but “checkpoint‑aligned” – the board cares about the next budget meeting.

Script excerpt (Ethan’s revised timeline email):

> Subject: Revised AI Delivery Schedule

> Board, per our July 5 call, the predictive‑readmission model will be in production by day 78 (July 22 2024). This meets the 84‑day board checkpoint and aligns with the $42 M Series B budget.

> 📖 Related: Costco PM promotion timeline leveling guide and review criteria 2026

Preparation Checklist

  • Review the startup’s latest Series B term sheet (e.g., Cerebra Health $42 M, 18 % discount) to extract cash‑flow milestones.
  • Map AI epics to the board’s OKR sheet (e.g., ScaleAI KPI #4 “Reduce latency to <200 ms”).
  • Calculate ARR impact per AI metric using the company’s finance model (e.g., FinTechX $2.3 M fraud reduction → $0.9 M net ARR).
  • Draft a 90‑day sprint Gantt that ends before the next board checkpoint (e.g., Helix Health 84‑day horizon).
  • Include a concise “Financial Impact” slide that shows dollar uplift per AI deliverable.
  • Work through a structured preparation system (the PM Interview Playbook covers “AI‑ROI framing with real debrief examples”).
  • Prepare a one‑page email template that mirrors the board’s communication style (e.g., subject line, concise bullet points, deadline).

Mistakes to Avoid

BAD: Submitting a technology‑first roadmap that lists “Transformer upgrade” without tying it to revenue.

GOOD: Presenting a “Transformer upgrade” that reduces inference cost by 12 % and adds $1.1 M ARR under the Series B profit target.

BAD: Using accuracy‑only metrics such as “99.9 % precision” that the board cannot translate to cash.

GOOD: Translating “0.5 % detection gain” into $2.3 M fraud reduction and $0.9 M net ARR, as shown in the FinTechX debrief.

BAD: Proposing a 12‑month timeline that overshoots the board’s next checkpoint.

GOOD: Aligning the AI launch to day 78 of an 84‑day board cycle, as Ethan Zhou did for Helix Health.

FAQ

Does a fractional AI head need to negotiate equity or salary?

Yes. At Cerebra Health, the board granted 0.35 % equity and a $185 k base salary after the candidate linked AI milestones to a $7.9 M ARR uplift. The equity figure is non‑negotiable unless the candidate can prove an additional $1 M ARR impact.

Can I use a generic AI proposal template for every Series B startup?

No. The ScaleAI debrief showed that a generic template that ignores the company’s cash‑flow calendar leads to a 6‑2 rejection. Each template must be customized to the startup’s specific OKRs, milestone dates, and finance model.

What is the ideal length for the AI portfolio proposal?

Three pages. The Helix Health candidate’s two‑page deck was rejected because it omitted a risk‑adjusted timeline; the three‑page deck that included a 90‑day Gantt received a 5‑3 vote for equity allocation.

---amazon.com/dp/B0GWWJQ2S3).

Related Reading

What does a Fractional Head of AI actually deliver for a Series B board?