Brand Graphics Landing Pages Blog Posts Case Studies In-house LLM prompt orchestrator + memory layer
FigJam playground

Workflow research — the in-house LLM playground.

Active FigJam board where the four agentic workflows are being mapped to our self-hosted LLM. Quick access to the live thinking.

Open in Figma

With agentic workflows — what we unlock

  • Compress time-to-draft. Repeatable assets go from hours to minutes across all four workflows.
  • Hold brand consistency. Every artifact looks like Intelligaia regardless of who briefed it.
  • Capture institutional knowledge. Visual and voice rules live in a versioned skill, not in private notes apps.
  • Free senior designer time. Heroes, pages, and standard case studies stop being a queue at the senior desk.
  • Compound improvement. Every correction sharpens the next run. The agents converge toward our taste.
  • Match AI-native cadence. Ship at the speed of studios using off-the-shelf AI — staying recognizably Intelligaia.

Without — what slows us down

  • Every asset starts from scratch. Designers re-prompt and re-iterate on work that should be repeatable.
  • Style drifts between designers. Brand consistency degrades quietly until a client notices.
  • Senior designers become bottlenecks. Every illustration and page needs a senior eye before it ships.
  • Onboarding takes weeks. Teaching our visual language to a new hire is a long look-and-mimic process.
  • Prompt knowledge leaks out. Private libraries leave with people. We rebuild from zero every hire.
  • Cadence lags. One or two blog heroes a quarter when AI-native studios ship daily.

Pick a workflow — see the three AI tools in action

Pilot
Planning

Brand Graphics — three AI tools in action

Hero illustrations and visual assets across five locked style modes. Pick the path that fits your access.

Pilot

Pure ChatGPT — setup walkthrough

One Custom GPT holds the style rules, banned-styles guardrails, and your reference images. You type a brief; DALL-E renders. ChatGPT Plus, Pro, or Team required (Custom GPTs aren't on the free tier).

1
Open the GPT Builder

Sign in at chatgpt.com. Left sidebar → Explore GPTs → top-right Create. Switch to the Configure tab.

2
Fill the Configure form
  • Name: Intelligaia Illustration
  • Description: "Generates illustrations in the Intelligaia visual language."
  • Instructions: Paste the contents of image-tool-instructions.md from the skill folder (~4,200 characters).
  • Conversation starters: Add four — one per mode (BL / YH / PP / SP / IT).
  • Knowledge: Upload 4–5 mode-prefixed reference images per mode (~20 total). Naming pattern: BL-fed.png, YH-head-fingerprint.png, etc.
  • Capabilities: Enable DALL-E Image Generation. Turn off Web Browsing and Code Interpreter.
3
Save and share

Top-right Create. Sharing scope: Anyone with the link for the team. Pin to your sidebar for one-click access.

4
Test it

Paste this test brief and confirm the output looks like the FED / GOODNESS / UX / W reference series:

Hero illustration for an upcoming blog post titled "AGENTS" — about the architecture of self-learning AI agents in design. BL mode. 16:9 hero.
Limit of this path. The GPT does not learn between runs. Every chat starts fresh with the same Instructions. If you find yourself making the same correction over and over, update the Instructions block — that's the only way the GPT "learns."

Landing Pages — three AI tools in action

Campaign, capability, partnership, and hire pages in Intelligaia tokens. Hero, value sections, social proof, CTA.

Planning

Pure ChatGPT — setup walkthrough

A Custom GPT loaded with the Intelligaia HTML token kit and 6–8 of our best published landing pages. You describe the campaign; the GPT returns a draft page in our tokens — hero, three value sections, proof grid, testimonial, CTA close. ChatGPT Plus, Pro, or Team required.

1
Open the GPT Builder

Sign in at chatgpt.com. Left sidebar → Explore GPTs → top-right Create. Switch to the Configure tab.

2
Fill the Configure form
  • Name: Intelligaia Landing Pages
  • Description: "Drafts landing pages in the Intelligaia design language using our locked HTML token kit."
  • Instructions: Paste the SKILL.md for intelligaia-landing-page — page-purpose options (campaign / capability / partnership / hire), structure (hero + 3 value + proof grid + testimonial + CTA), token rules (cream, ink, yellow as accent), banned filler.
  • Conversation starters: Four — one per page purpose: "Draft a campaign LP for [launch]", "Draft a capability LP for [service]", "Draft a partnership LP for [partner]", "Draft a hire LP for [role]".
  • Knowledge: Upload the Intelligaia HTML token kit (one file with header, footer, section catalog, color tokens, type scale) plus 6–8 best published landing pages as HTML reference.
  • Capabilities: Web Browsing optional. DALL-E and Code Interpreter off — visuals farmed out to Brand Graphics; this GPT does layout and copy.
3
Save and share

Top-right Create. Sharing scope: Anyone with the link for the team. Pin to your sidebar for one-click access.

4
Test it

Paste this brief and confirm the GPT returns a structured page in Intelligaia tokens (cream, ink, yellow accent — no lorem ipsum):

Draft a campaign LP for our agentic AI workshop launch.
Audience: design directors.
Primary CTA: book a discovery call.
Proof points: HAX, Atlas, AI-EAM.
Standard 6 sections.
Limit of this path. The GPT does not learn between runs. Recurring corrections — a CTA verb that always gets rewritten, a section that always gets cut — need manual Instruction updates.

Blog Posts — three AI tools in action

Drafts in the Intelligaia voice from a topic and optional source material. Hook, tension, POV, evidence, close.

Planning

Pure ChatGPT — setup walkthrough

Custom GPT seeded with our voice glossary, banned-filler list, and five reference posts. Drop a topic, a link, or some notes; the GPT returns a draft in our voice plus three title variants and a meta description. ChatGPT Plus, Pro, or Team required.

1
Open the GPT Builder

Sign in at chatgpt.com. Sidebar → Explore GPTs → top-right CreateConfigure tab.

2
Fill the Configure form
  • Name: Intelligaia Blog Posts
  • Description: "Drafts on-brand Intelligaia blog posts — title, hook, structure, POV, CTA — in our voice."
  • Instructions: Paste the SKILL.md for intelligaia-blog-post — voice glossary, structure (hook → tension → POV → 2–4 evidence sections → counter-position → action close), banned filler ("leverage", "seamless", "unlock", "at the end of the day"), output spec (3 title variants + meta description + hero illustration brief).
  • Conversation starters: Four — "Draft a POV post about [topic]", "Turn this talk into a post", "Write up our perspective on [trend]", "Convert these notes into a blog post".
  • Knowledge: Voice glossary + 5 best published posts in markdown.
  • Capabilities: Web Browsing on. DALL-E off — hero illustration handoff goes to Brand Graphics.
3
Save and share

Top-right Create. Sharing: Anyone with the link for the team. Pin to your sidebar.

4
Test it

Paste this brief and confirm the draft hits the voice, structure, and length:

Draft a 1200-word POV post titled
"Why agentic design needs a human in the loop."
Reader: design directors.
Source: notes from our last team retro.
CTA: book a consult.
Limit of this path. The GPT can't learn between runs. If the same paragraph-level rewrite keeps happening — cutting the 5-paragraph intro, swapping abstract claims for one concrete example — update the Instructions block.

Case Studies — three AI tools in action

Portfolio writeups from project material. Hero, problem, approach in three stages, metrics, testimonial.

Planning

Pure ChatGPT — setup walkthrough

Custom GPT with the Intelligaia portfolio HTML template baked into Knowledge. Drop client materials (brief, screenshots, outcome notes); the GPT returns a portfolio-ready draft with NDA-aware outcome phrasing and screenshot crop suggestions. ChatGPT Plus, Pro, or Team required.

1
Open the GPT Builder

Sign in at chatgpt.com. Sidebar → Explore GPTs → top-right CreateConfigure tab.

2
Fill the Configure form
  • Name: Intelligaia Case Studies
  • Description: "Turns finished project material into a polished Intelligaia case study in the portfolio HTML style."
  • Instructions: Paste the SKILL.md for intelligaia-case-study — portfolio structure (hero + problem + approach in 3 stages + highlight screen + metrics + testimonial + next-engagement CTA), NDA outcome-phrasing rules, client-team crediting patterns, banned claims (never invent metrics, never overclaim).
  • Conversation starters: Four — "Case study for [client project]", "Write up the [client] engagement", "Turn these notes and screenshots into a case study", "Portfolio page for our work with [client]".
  • Knowledge: Intelligaia portfolio HTML template + 3 best published case studies + NDA phrasing guide.
  • Capabilities: Web Browsing optional. DALL-E off — screenshots come from the project file, never from generation.
3
Save and share

Top-right Create. Sharing: Anyone with the link for the team. Pin to your sidebar.

4
Pilot with one historic project

Pick a project that's already shipped (so you know the right answer) and re-write it. Compare against the actual published version to calibrate. Starter brief:

Case study for [client].
Sector: enterprise SaaS.
Problem: scaling design ops across 4 product lines.
Approach: shared design system + research ops pod.
Outcomes: 3× case study velocity, NDA on revenue impact.
Materials: 18 Figma screens + 2 client testimonials + outcome memo.
Limit of this path. The GPT can't learn between runs. NDA phrasings and client-team crediting patterns will need to be re-tightened in Instructions over time.