DEV Community: Axiom Lab

Your Demo Day Video Shouldn't Be a Slide Read-Along

Axiom Lab — Tue, 13 Oct 2026 22:03:34 +0000

Founders treat the demo-day cut like a compressed deck: title, problem, solution, market, ask—spoken over the same bullets the judges already have in PDF. That version gets watched at 1.5x and forgotten. The clip that sticks is the one that shows why you exist in under ninety seconds, then points people somewhere they can dig deeper.

Here is a structure that works when the room is tired and the next founder is already lining up.

Open on the moment of pain, not the company name

Skip the logo sting. Start on the concrete failure: the founder who spent three nights rebuilding the same investor intro, the ops lead reconciling three tools by hand every Monday, the recruiter whose LinkedIn pitch gets ignored by senior engineers. Name the person and the friction. If a stranger can't picture it in eight seconds, they won't care about your category later.

Show the before/after in one beat

Demo day isn't a feature tour. One clear before, one clear after, filmed or storyboarded so the change is visible without narration. If you need a voiceover, keep it under two sentences. The video should still make sense on mute with captions.

Say the ask once, then stop

"We're raising a seed to ship X to Y customers" beats a laundry list of milestones. Say it once near the end. Don't reopen the problem. Don't apologize for the ask. Cut.

Leave a durable link, not a QR that dies after the event

Judges, angels, and operators who couldn't stay will Google you later. Put a stable URL on the last card—something like axiomlab.online—so the video keeps working as a founder card long after demo day. That same clip can then sit in cold DMs, accelerator forms, and weekly updates without remakes.

What to cut

The market-size slide recited aloud
Team bios that belong in a one-pager
Any shot of you reading bullet points off a monitor
A soft close that trails into "happy to chat anytime"

A ninety-second checklist

Second 0–8: who hurts and how
Second 8–45: the change, shown not claimed
Second 45–70: why now / why you (one proof)
Second 70–90: the ask + where to go next

If you need a pipeline that turns a short script and storyboard into a finished founder video with QA gates before anyone sees a half-broken render, that's the problem we built Axiom Lab around. Use the checklist above either way—the format matters more than the tool.

Demo day is loud. Make the ninety seconds the quiet part that still gets remembered.

Build One 75-Second Founder Card (Then Stop Remaking Intros)

Axiom Lab — Tue, 13 Oct 2026 10:35:33 +0000

Most early teams do not have a video problem. They have an intro-rewrite problem.

Every new channel asks for the same story in a slightly different box:

Cold email to an investor
Accelerator application "founder video"
Warm intro forwarded to a partner
Hiring page "meet the founders"
Conference speaker bio that nobody will click

So the founder opens a blank doc, writes a new paragraph, records a new take (or skips recording again), and ships something slightly worse than last month. The story drifts. The CTA changes. The deck and the spoken pitch disagree.

A Founder Card is the opposite approach: one reusable 60–90 second intro film, versioned on purpose, linked everywhere the long story used to be rewritten.

This post is a concrete build guide. No vanity metrics. Just a structure you can finish this week.

What a Founder Card is (and is not)

Is: a short decision aid. After watching, a stranger should know who you are, what you sell, who it is for, and what to do next.

Is not: a brand anthem, a feature tour, a fundraising montage, or a substitute for your deck. The deck still carries depth. The card carries orientation.

Think of it as the spoken equivalent of your one-liner — with a face, a scene, and a single next step.

The four beats (keep them in this order)

1) Identity (0–10s)

Face on camera (or a consistent avatar), name, role, company, and the product in one sentence.

Bad: "We're building the future of AI-powered workflows for modern teams."
Better: "I'm [Name], co-founder of [Company]. We help early founders turn a deck and a short interview into an investor-ready pitch video."

If the product is missing from sentence one, the rest of the card is recovery mode.

2) Scene, not market size (10–35s)

Describe a moment the viewer already recognizes. Numbers can wait.

Example scene: "You attach a 30-page PDF to a cold email. The partner opens it on a phone between meetings. The file becomes homework, not a conversation."

One vivid scene beats three TAM slides spoken aloud.

3) Mechanism in plain words (35–55s)

Say what the customer submits and what they get back. Skip the model names.

Example: "You upload the deck, talk through a short AI-hosted interview, and get a pitch cut you can send. You pay when a delivered version passes QA — not for failed renders."

Input → output → when money changes hands. That is the product for a cold viewer.

4) One next action (55–75s)

Exactly one:

Book a 15-minute fit check
Reply with "deck" and get a review checklist
Open the sample page and watch a finished card

Two CTAs cancel each other. Pick the action that matches where you will paste this link.

Production rules that keep quality high without a studio

Audio first. A quiet room beats a pretty background. Phone mic 20–30cm from your mouth is fine.

Short sentences. Especially if English is not your first language. One idea per sentence. Native fluency matters less than clarity.

No slide reading. If you catch yourself narrating bullet points, stop and rewrite the beat as a scene.

Version labels. Call files founder-card-v1-angel-cold.mp4, v2-accelerator.mp4. Same spine, different final line. Do not pretend one cut fits every audience forever.

Thumbnail. Face or product UI plus ≤8 words. Black frames and final_REAL_v7.mp4 filenames kill opens.

Where to place the same card (without sounding spammy)

Channel	How to use it
Cold email	Link in line 3 of a 5–7 line note; PDF optional, not centered
Accelerator form	Upload the accelerator-CTA version; keep under their time cap
Warm intro	Paste link so the forwarder does not have to re-explain you
Hiring	About page or Jobs pin — candidates want a human, not a slogan
Partner deck	First slide QR or short URL; keep the spoken CTA partner-shaped

The rule: same spine, swapped last line. Remaking the whole story for each channel is how teams stall for months.

A 45-minute v1 checklist

Write four bullets only: identity / scene / mechanism / CTA.
Time a read-aloud. Cut until it lands between 60–90 seconds.
Record three takes; keep the clearest audio, not the "perfect" face.
Export, add a simple title card with product name + URL.
Send to one friendly operator and ask: "What would you do next?" If they shrug, fix the CTA beat before polishing B-roll.

Ship v1. Collect the first three objections from real replies. Fold those into v2. Perfection is how Founder Cards never leave the drive.

When filming is the bottleneck

Some teams cannot put a co-founder on camera this month: split cities, heavy accents under stress, application season with weekly script changes.

In those cases the bottleneck is not "better lighting." It is re-running the same pipeline when inputs change. That is the problem we built Axiom Lab around: deck + AI-host interview in, pitch video / Founder Card out, charge only after QA on the delivered film. Public entity: 深圳溪有木科技有限公司. Press: sales@axiomlab.online.

Whether you film yourself or use a pipeline, the editorial bar stays the same: four beats, one CTA, versioned on purpose.

Takeaway

Stop treating every new intro as a blank page. Build one Founder Card, label the version, swap only the last line per channel, and retire the habit of rewriting your origin story every Friday night.

If you already have a card, reply or comment with the CTA line you use most — the weak endings are usually where good films die.

Your Pitch Deck Is Not Your Pitch Video

Axiom Lab — Tue, 13 Oct 2026 02:32:29 +0000

Founders keep asking AI video tools to "turn this deck into a video." The results are usually a slideshow with a voiceover that reads the bullets aloud. That is not a pitch. It is a worse deck.

A usable pitch video has to do three jobs a slide stack does poorly: establish who is speaking, make one problem feel real in under a minute, and end on a clear next step. Decks are optimized for skim-and-debate in a meeting room. Videos are optimized for attention under uncertainty — an investor watching on a phone between calls, a partner deciding whether to take the meeting at all.

This post is a practical workflow for bridging that gap without producing another talking-slideshow. It is the same workflow we use at Axiom Lab when we turn decks, PDFs and a short AI-host interview into a pitch film or a short "Founder Card" introduction.

Start from the audience's first 30 seconds

Before you touch any generation tool, write down who will watch this and what decision they are making. Three common cases:

Cold intro — they do not know you. Goal: earn a reply or a meeting.
Warm follow-up — they already talked to you. Goal: refresh the story and lock the ask.
Internal alignment — co-founders, advisors, early hires. Goal: one shared narrative.

If you cannot name the decision, you will generate a generic "about us" film. Generic films get skipped.

For cold intros, the first 30 seconds must answer: who are you, who do you help, and why should I keep watching. Decks often bury that under market size slides. Videos cannot.

Extract the story; do not narrate the slides

The failure mode is treating each slide as a scene. Slide 3 becomes scene 3; the voice reads the title; you cut to the next title. Viewers hear a table of contents.

Instead, pull four beats out of the material and ignore the rest for the first cut:

Identity — who is speaking and why they are credible on this problem.
Pain — one concrete situation a customer was stuck in.
Change — what is different after your product or service.
Ask — the single next step (book a call, try a pilot, reply with a date).

Everything else (TAM, competitor matrix, full roadmap) can stay in the PDF attachment. The film's job is to make someone open that PDF with intent.

A useful test: if you muted the video, would the on-screen text still tell a story, or only list nouns? If it only lists nouns, rewrite the beats before you render.

Use an interview to get spoken language, not brochure language

Founders are usually bad at writing about themselves and good at answering questions. A blank script box produces "We are a passionate team leveraging AI to disrupt…" An interviewer asking "Who was the first customer, and what changed for them on Tuesday?" produces a sentence you can actually put in a film.

That is why our Founder Card flow runs a short AI-host interview before production. The interview is not a gimmick. It is an input-quality tool. Better spoken answers mean fewer script revisions and fewer "this does not sound like me" re-renders later.

If you are doing this by hand, record yourself answering five questions out loud, then transcribe:

Who do you help in one sentence?
What were they doing before that was painful?
What do they do now that is different?
What proof can you show without inventing numbers?
What should a viewer do in the next 24 hours?

Cut filler. Keep names, concrete verbs, and one proof point you can stand behind. Do not invent metrics or customers for the script — viewers and platforms both punish that, and you will have to walk it back in the first call.

Freeze a plan before you spend on renders

AI video is expensive when you iterate blindly. The cheapest place to iterate is a written plan the founder can approve: beats, claims, and which source document supports each claim.

Only after that plan is frozen should you generate picture, voice and captions. If the story is wrong, no amount of prettier frames will save it. If the story is right, a failed render is just a technical retry.

This is also why pay-on-delivery pricing forces better product design: when you only charge for a film that passes QA, you cannot afford to discover the story in the render loop. Approval gates are cost control, not just UX.

Keep claims tied to sources

Pitch videos get people in trouble when they overclaim. A simple discipline:

Every factual claim in the narration should map to a line in the deck, a document, or something the founder said in the interview.
Prefer "we work with early-stage B2B teams on X" over vague superlatives.
Put numbers only when they already appear in approved materials.

If a line cannot be sourced, cut it. Short and true beats long and shiny.

Structure that works for a short Founder Card

For a short introduction film (roughly a minute, not a full pitch), a tight structure is enough:

Name and role on screen.
Who you help and the pain in one breath.
What you deliver and one proof or experience point.
Contact or clear next step.

That is the Founder Card shape we use at Axiom Lab: identity, service value, experience and proof, contact. It is deliberately small so the interview has a finish line and the film has one job — make a stranger willing to talk.

Longer project videos can expand the middle (problem depth, product walkthrough, traction), but they should still open and close the same way. Do not save the ask for "somewhere in the description."

Production checklist before you hit generate

Audience and decision named.
Four beats written in spoken language.
Ask is a single verb plus destination (email, calendar link, site).
Claims sourced; no invented metrics.
Aspect ratio and length match how it will be sent (email vs. LinkedIn vs. site embed).
Captions planned — many people watch muted.

Then generate. Watch once for story, once for technical issues (names mispronounced, wrong claim, missing CTA). Fix the plan if the story is wrong; retry the step if the story is right and the render failed.

Where Axiom Lab fits

If you want this pipeline without building it yourself, Axiom Lab takes a founder's deck and materials, runs a short AI-host interview, produces a plan you approve, then delivers a pitch video or Founder Card. Charging is pay-on-delivery for films that pass QA — failed jobs do not keep the hold.

Public entity: 深圳溪有木科技有限公司. Press and sales: sales@axiomlab.online.

The tool is optional. The discipline is not. Decks explain; videos persuade under time pressure. Treat them as different artifacts, extract a spoken story first, freeze the plan, and only then spend on frames.

Building a Founder-Card Video Pipeline: From AI Host Interview to Delivery Gate

Axiom Lab — Mon, 12 Oct 2026 12:54:05 +0000

Founder intro videos fail for the same reason pitch decks fail: we jump to output before we lock structure.

This post walks through a builder-friendly pipeline for AI-assisted founder-card videos — the kind where an AI host interviews you, and the script, look, and voice come out of that conversation — plus the production gates that keep quality from collapsing when generation gets cheap.

If you’re shipping video features, running an agency-lite workflow, or just tired of teleprompter takes, treat this as a reference architecture.

What a “founder card” actually is

Think of a compact introduction film with four fields:

Field	Job
Identity	Who you are / how you’re positioned
Service value	Who you serve and what changes for them
Experience & proof	Evidence you’ve done the work
Contact	How to collaborate next

It’s not a brand reel. It’s a portable personal PRD in video form — useful for partner intros, inbound reply links, accelerator profiles, and sales follow-ups.

Pipeline overview

capture → structure → plan freeze → generate → QA gate → deliver (then charge)

The founder-card pipeline: interview → script, look & voice → checks → delivery.

Most DIY stacks stop at generate. The interesting product work sits in structure, freeze, and QA.

1) Capture: interview beats script

A short AI-hosted interview outperforms a blank script editor for three reasons:

Founders answer questions more naturally than they write copy.
Follow-ups surface proof (“what shipped?”, “who was it for?”) you’d forget to type.
The transcript becomes a structured source of truth — not vibes.

Minimal capture contract:

Name + positioning one-liner
Audience
Value proposition
1–3 proof points
Preferred CTA / contact

Optional: upload a PDF bio, LinkedIn export, or past talk slides as grounding context.

2) Structure: map answers → beats

Convert the interview into a beat sheet before any render:

Hook (identity in one breath)
Who you serve
Value in a concrete scenario
Proof
Soft ask / contact

Keep each beat to one sentence in the plan. If a beat needs a paragraph, it needs a cut — not a longer voiceover.

3) Plan freeze: the PRD for the film

Before generation, produce a human-readable production plan:

Ordered beats
On-screen evidence (logo, metric, screenshot) per beat
Tone (peer / investor / client)
Hard exclusions (no feature dump, no 4-minute cut)

Freeze means: production starts only after approval. This is the difference between “AI video toy” and “dependable manufacturing.”

4) Generate: treat models as workers, not directors

Generation can include:

Voice synthesis from the approved script
B-roll / still sequencing from uploaded photos or deck pages
Lower-thirds for name and CTA
Pacing aligned to beat boundaries

Keep directors’ rights with the plan. Models propose cuts; the frozen plan arbitrates.

5) QA gate: fail closed

Define automated + human checks before delivery:

Duration within target band
Every beat present in order
CTA audible and on-screen once (not five times)
Proof assets match claims (no hallucinated logos)
No unresolved placeholders

Fail closed: if checks fail, don’t deliver and don’t charge. That commercial rule is a product feature — it forces the pipeline to be honest about readiness.

20-second walkthrough of the three steps.

Why “pay on delivery” belongs in the architecture

Billing only after a checked delivery isn’t just nicer packaging. It changes system design:

You can’t hide behind deposits for half-finished work.
QA becomes revenue-critical, not optional polish.
Founders trust the loop enough to upload messy real materials.

If you’re building adjacent tooling, mirror the same gate even if pricing differs: no public artifact without a passed checklist.

Billing sits after the QA gate, not before production.

Reference checklist (copy/paste)

Inputs

[ ] Interview complete (or transcript uploaded)
[ ] Audience + CTA chosen
[ ] At least one proof asset available

Plan

[ ] 5–7 beats, one sentence each
[ ] Ordered evidence map
[ ] Explicit out-of-scope list
[ ] Founder approved freeze

Output

[ ] Duration OK
[ ] Beat order verified
[ ] Claims backed by assets
[ ] Contact/CTA correct
[ ] Internal QA signed off → deliver

Where Axiom Lab fits

Axiom Lab runs this loop for pitch/project films and founder cards: upload plans, decks, PDFs, or photos (or complete an AI host interview for a founder card), agree the production plan, then receive a film only after it passes checks — charged on delivery.

Builders can steal the pattern even if you never use the product: interview → structure → freeze → generate → QA → deliver. Generation is the easy part now. Trust comes from the gates.

Next experiments worth shipping

Structured interview schemas (JSON) shared between capture and editor
Diff view when a founder revises a frozen plan mid-flight
Per-audience plan variants from one interview (investor vs. customer cut)
Open example repos: sample beat sheets + QA rubrics (great GitHub README material if you open-source the non-secret parts)

If you build video features, invest in the boring middle — plans and gates — not only the flashy render step. That’s how AI-assisted production stops looking like a demo and starts feeling like a pipeline.

Shipping founder intros or pitch films? Start with materials you already have at axiomlab.online.