Yesterday (March 26. 2026) I attended Kaila’s (The AI Homegirl’s) app-building workshop at CoBiz, and we did something I honestly didn’t expect to pull off in a single session: we built and published real, live personal brand websites — no code required.

The platform was Lovable.dev, a no-code site builder powered by AI. The workflow combined ChatGPT for prompt development with Lovable for actual site generation. Two hours later, every participant had a published landing page with a live URL. Here’s everything I took away.

---

## What Makes a Good Personal Landing Page

Before touching any tools, Kaila walked us through the essential “ingredients” of a strong personal brand site:

- **Hero section** with your name and up to 3 related professional titles

- **About section** with a bio, headshot, and visual storytelling

- **Offers/services** with clear value propositions

- **Social proof** — testimonials, certifications, awards, press

- **A clear call-to-action** for contact or booking

One key insight: lead with your strongest credentials immediately. Visitors decide whether to stay within seconds, so don’t bury the lede. As Kaila put it - make it social proof!

We workshopped this live with my info, and I was encouraged to include some of my career and personal highlights: including 7 patents, a published UX research book chapter, a “Tech Impact Award” from Google, and a children’s book honor from the International Board on Books for Young People (IBBY) organization. The exercise of identifying *which* credentials to lead with — and in what order — was itself valuable.

---

## How the Build Process Works

Lovable charges credits per prompt interaction, so the workflow is designed to minimize unnecessary back-and-forth inside the platform. Here’s the basic flow:

1. Develop your full prompt using the **Lovable GPTs on ChatGPT** (free)

2. Transfer the optimized prompt into Lovable

3. Generate your site (~1–2 credits for a full landing page)

4. Publish immediately to a free `lovable.app` subdomain

5. Refine, then connect a custom domain when ready

At $25/month for 100 credits, being strategic about how you use them matters.

---

## Kaila’s 8 Tips for Getting the Most Out of Lovable

**1. Use ChatGPT first to save credits**

Build out your full prompt in the Lovable GPTs on ChatGPT before touching Lovable. Since Lovable charges per prompt, fleshing out your idea for free in ChatGPT first means fewer back-and-forth iterations in Lovable.

**2. Use Plan Mode for complex features**

Plan Mode costs a flat 1 credit regardless of complexity, and it won’t touch your code. Use it to map out how to implement something — like connecting a Google Business Profile — before committing to a build.

**3. Be as specific as possible in your prompts**

Vague prompts leave room for unexpected results. The more specific you are about sections, tone, and content, the more consistently you’ll get what you’re after.

**4. Lead with emotional direction**

Lovable responds well to tonal guidance like “warm and friendly” or “bold and visionary.” This informs design choices holistically — not just colors and fonts, but layout, imagery style, and overall feel.

**5. Use the Lovable GPTs, not regular ChatGPT**

The Lovable-specific GPTs are trained on Lovable’s documentation, so they generate prompts far more optimized for the platform than generic ChatGPT would produce.

**6. Publish early with the free subdomain**

You get a `lovable.app` subdomain immediately. Publishing early lets you test on mobile and share for feedback before you ever connect a custom domain.

**7. Track your credit usage in Settings**

Check your project settings to monitor how many credits you’ve used. A full landing page should cost ~1–2 credits; individual refinements should cost less.

**8. Use Flexy.co for style prompts**

This tool (built on Lovable) lets you pick a visual style and generates an optimized prompt with your colors and preferences — great for quickly redirecting your site’s aesthetic.

---

## A Note on UX Philosophy (my contribution to the group)

Beyond the technical tips, Kaila wanted to ensure to weave in some foundational UX thinking and asked me to give an overview of user experience and what to think about when building anything. Good user experience isn’t just about making something functional — it’s about matching people’s mental models, offering moments of delight, and making interactions feel effortless. One of the workshop’s running examples was a Kaila’s own creation - the “Homegirl Mode” translation feature: a small personality touch that makes an interface feel distinctly *AI Homegirl* rather than generic.

That framing — **personality as differentiation** — applies whether you’re building a landing page, a product, or a brand.

---

## What We Built

By the end of the session, two of us who participated had live, published sites (landing pages):

- **Aaliyah’s site**: A beauty educator focus with a warm, friendly aesthetic

- **Lulit’s site** (at https://lulit.lovable.app/): A bold, visionary design emphasizing my innovation and research background

Both were generated in under an hour and published before the workshop ended.

If you’ve been putting off your personal brand site because it felt too technical or too time-consuming, Lovable genuinely lowers that barrier. The combination of thoughtful prompting and a platform built for iteration makes it more accessible than anything I’ve tried before.

---

*Workshop hosted by Kaila (AI Homegirl) Love at CoBiz, March 2026.*

🧠 Embodied Futures: Innovating for Life — Monika Bielskyte

Monika Bielskyte joined us virtually from South Africa

The second part of the conference began with a presentation from Monika Bielskyte, futurist at Nike, who joined us virtually from South Africa. She opened with:

“Future happens to our bodies.”

Her talk, Embodied Futures, explored how “our future is increasingly shaped by big disinformation” and that “those who control fantasy control the future.” She cautioned that dystopia breeds despair and escapism, and urged us to:

“Follow the science and not the hype.”
“Not every technology has this magical element. Technology is not magic. If the problem is social, cultural, then the solution has to be as well. #NerdRapture”

She shared six Protopian Principles, describing a process rather than a destination:

• Plurality – Beyond Binaries

• Interdependence – Beyond Borders

• Biosphere Regeneration

• Sensory Expansion / Joy!

• Creative Flourishment

• Spiritual Fulfillment

Her concept of life-centric design, “anchored in the scientific understanding of interconnectedness,” left us reflecting on our shared responsibility:

“We are in this together.”

Slide showing Monika’s six Protopian Principles

⚖️ Embedding Social Justice & Equity in Model Building — Donald Martin Jr.

AI Development Cycle by Donald Martin Jr. presenting on Societal Context (SCOUTS) tools

Donald Martin Jr., tech executive and Responsible AI expert, delivered a masterclass on understanding societal context when developing AI.

He began with a personal story about his mother’s healthcare journey to illustrate how “assumptions humans make based on surface-level concepts impacts us.”

Donald walked us through the AI development lifecycle, emphasizing that:

“Most important in any development is understanding the problem and not data collection.”

He reminded us of Einstein’s lesson:

“If I had an hour to solve a problem, I’d spend 55 minutes thinking about the problem and five minutes thinking about solutions.”

His takeaway:

“Be like Einstein. Prototype your problem before thinking about solutions.”

He also recommended Thinking in Systems by Donella H. Meadows as essential reading for all AI practitioners.

🤖 Desinging Human-Centered AI — Carol J. Smith (Carnegie Mellon University)

Carol J. Smith, Principal Research Scientist at the Software Engineering Institute (SEI) at Carnegie Mellon University, closed Day 1 with a presentation on developing calibrated trust in AI systems.

She described our “need for automation and desire to make things easy,” but warned that we often fall for “illusions of understanding and of objectivity.” We “buy into hype,” and must be intentional in choosing the right tools.

“Trust is personal and we calibrate based on personal experiences, current context, and available evidence of the system’s capability and integrity.”

She closed with timeless advice:

“Be skeptical and activate curiosity.”

💬 Q&A Panels to Wrap Up Day 1

Q&A panel with Dr. Jamika Burge, Ruth Kikin-Gil, and Tarzine Jackson

To conclude the day, Dr. Jamika Burge moderated two Q&A sessions — first with Ruth Kikin-Gil (Microsoft) and Tarzine Jackson (Koalesce Designs), then with Ovetta Sampson and Dr. Chinasa T. Okolo (Brookings Institution and Technēcultură). These conversations offered insightful reflections and helped cement a full day of learning.

Group Photo

🧩 Day 2 – Hands-On Workshops

The final day featured three interactive workshops that translated the summit’s themes into practice.

🧱 Workshop 1 — Reverse Engineering AI Design to Identify and Minimize Human Engagement Risks (HER)

Facilitator: Ovetta Sampson

Ovetta introduced core concepts — ML, AI, UX/UI Design — then moved into Designing Trustworthy AI and Deconstructing AI.
Her session was “so jam-packed with golden nuggets” that it “could easily be taught over the course of a few days.”

“I would highly recommend taking this workshop if you can, and I hope to attend more of Ovetta’s courses in the future.”

Ovetta Sampson facilitating her workshop

🌐 Workshop 2 — Understanding the Global Impacts of AI to Advance Novel Inquiry

Facilitator: Dr. Chinasa T. Okolo

Dr. Chinasa Okolo guiding participants through worksheets

Dr. Okolo shared global AI data, asked participants to formulate their own AI-related research questions, and guided us through defining scope, methods, community, and impact.

“The worksheet provided and discussions really helped us get some hands-on experience that we can leverage for our own work.”

🎨 Workshop 3 — Designing AI from the Inside Out

Facilitator: Paz Perez

Paz Perez leading Designing AI from the Inside Out workshop

Paz led a dynamic session on configuring Gemini (GenAI) for UX and research. She demonstrated the difference between good and bad prompts, helped us design our own, and test them live.

“It was super helpful to use these examples and the framework she provided to ensure how to guide and define the way GenAI would respond to our chat prompts.”

💡 Final Reflections

From Monika Bielskyte’s call to “follow the science and not the hype”, to Donald Martin Jr.’s reminder to “prototype your problem before thinking about solutions,” and the hands-on sessions from Ovetta, Chinasa, and Paz — every moment underscored a single theme:

Responsible AI is built through understanding context, embracing diversity of thought, and designing with intention and care.

As the summit came to a close, I left with renewed conviction that our work in UX and AI must always start with people — their cultures, their stories, and their lived realities.

#ReimaginingAI #BlackComputeHER #AIForGood #EthicalAI #InclusiveDesign #UXResearch #HumanCenteredAI

This first reflection covers the opening reception at Spelman College and the first half of Day 1 at the Woodruff Art Center — before diving into the design sessions and workshops that will appear in Part II.

🏛 Pre-Summit Reception — Spelman College

Hosted by Dr. Jamika Burge of BlackComputeHER.org and Ovetta Sampson, the evening gathered speakers, founders, and thought leaders inside the halls of the historic HBCU.

Among the special guests was Christopher Priest, legendary writer of The Black Panther. Meeting him was an honor — a reminder that storytelling and technology both shape how we imagine the future.

Meeting the writer of The Black Panther, Christopher Priest — a highlight of the evening.

I also connected with several innovators driving inclusive AI:

• Chad Reynolds (Vurvey Labs)

• Tarzine Jackson (Koalesce Designs)

• Peta Gaye Clarke (Women Tech Exchange)

• Gerald Carter (Destined AI)

• Cindy Causey (Visa)

• and my former Google colleague Paz Perez, who would later lead a hands-on generative AI workshop.

  

  With Paz Perez (left), Dr. Jamika Burge (center), and Tarzine Jackson (right) — three women leading the charge in inclusive AI design.

The evening was full of rich conversation — from future collaboration ideas to reflections on how AI can center human values.

Meeting Dr. Jamika Burge in person for the first time — a moment of mutual admiration and shared purpose.

It was an inspiring start — a space filled with warmth, intellect, and anticipation for what lay ahead.

🎙 Day 1 — The CEO Panel

The summit’s first morning at the Woodruff Art Center opened with a thought-provoking CEO panel moderated by Ovetta Sampson.

Panelists:

• Riana Lynn, Journey Foods

• Chad Reynolds, Vurvey Labs

• Gerald Carter, Destined AI

• Vivianne Castillo, Choose Courage Inc.

• Vincent Braithwaite, /gidens/
  

Each founder shared their “why” — the personal motivations and social contexts behind their companies:

• Riana Lynn spoke about building equitable access to healthy food through data-driven insights at Journey Foods: “to build the most actionable database.”

• Chad Reynolds traced his journey back to childhood product-testing sessions with his mother at P&G:
  “If you want to build anything for this world, it has to incorporate people and feedback.”

• Gerald Carter revealed how biased image-recognition systems pushed him to found Destined AI, striving for fairness and representation.

• Vincent Braithwaite explored how AI can streamline workflows while preserving empathy:
  “Moving fast doesn’t mean you care less. The age of AI demands speed and discernment.”

• Vivianne Castillo described founding Choose Courage Inc. to champion human-centered entrepreneurship and well-being — even launching Moonpaws, “astrology for pets,” to remind us that empathy matters in all forms.
  

Ovetta’s questions went deep — examining ethics, intentionality, and leadership:

“How do you decide to use technology that has been used as a weapon against marginalized communities?”
“When do you pass on AI because of your values?”
“How do we balance speed and accuracy without losing humanity?”

A few memorable quotes lingered:

“Be transparent and open — and have difficult conversations.” — Vincent Braithwaite
“Creativity is not a solo sport. Build agents to help you unlock your creativity.” — Chad Reynolds
“Be data-informed, not data-driven.” — Vincent Braithwaite

Morning CEO panel at the Woodruff Art Center, moderated by Ovetta Sampson (left) with Riana, Chad, and Gerald on stage. Vincent and Vivianne joined virtually.

🌍 Ubuntu Design Thinking — African Intelligence for AI

The morning closed with Wakanyi Hoffman, an Indigenous Knowledge Scholar and Ubuntu Philosopher, delivering “The Business of Ubuntu: How to Ethically Engage with the Community for AI Innovation in Africa.”

She urged technologists to move beyond the “dark-continent” myths that still bias global AI systems and instead design from Africa’s relational worldview.

“Ubuntu is relational intelligence — I am because we are.”

Her call to action:

• Ask “Why build?” before “How to scale.”

• Model humans with others and with the land.

• Build AI in an African image — AI for AI.

She closed with a Kenyan saying that lingered with me:

“Medua Aiangak — Those who live together can’t see each other; they are one.”

💡 Reflections

From the historic campus of Spelman to the art-filled halls of Woodruff, Part I of the summit revealed a powerful truth:

AI must reflect human intelligence — ethical, cultural, and collective — as much as computational power.

Part II will explore the second half of Day 1 and the design and research workshops on Days 2, where creativity meets practice.

Stay tuned.

#ReimaginingAI #BlackComputeHER #AIForGood #EthicalAI #InclusiveDesign #UXResearch #HumanCenteredAI

Across three days (October 23–25), the summit will explore accessible, equitable, and forward-looking AI design — topics that are deeply aligned with my own work and passions.

The sessions promise to stretch how we think about the evolving relationship between humans and intelligent systems:

• Developing Accessible & Futuristic AI — accessibility not as an afterthought but as a design principle.

• Designing for AI Experiences — how we shape trust, transparency, and collaboration between humans and AI.

• AI User Research Panel — a highlight for me; I’m curious how others are redefining research for adaptive systems.

• Workshops on Intelligent System Design — shifting from interface design to AI behavior design.

The Speakers

The lineup reflects a powerful mix of thought leadership and practice:

• Ovetta Sampson and Jamika D. Burge – reminding us that equity must be embedded, not added on.

• Ruth Kikin-Gil, Carol J. Smith, and Chinasa Okolo – bringing critical perspectives on ethics, responsibility, and governance.

• Monika Bielskyte – expanding our imagination through speculative futures.

• Vincent Brathwaite, Chad Reynolds, Tarzine Jackson, and Paz Perez – reimagining design for an intelligent age.

• Vivianne Castillo and Riana Lynn – grounding innovation in empathy and courage.

• Donald Martin Jr. and Gerald Carter – ensuring leadership and entrepreneurship drive responsible AI.

I am also particularly interested to hear from Wakanyi Hoffman, founder of The African Folktales project, speaking on integrating Ubuntu Ethics into AI systems design thinking.

What I’m Bringing to the Summit

As a UX Researcher passionate about AI-driven experience design, I’m attending with three intentions:

1. Learn how peers are addressing inclusion, bias, and accessibility in real-world AI systems.

2. Connect with practitioners and educators driving the future of responsible tech.

3. Reflect on how human-centered research can influence the next generation of AI design tools and platforms.

I’ll be sharing notes and reflections after each day here on Substack — highlights, quotes, and takeaways on how we can design AI that uplifts rather than marginalizes.

Stay tuned for Day 1 reflections soon after the opening session!

After trying out Replit a few weeks ago, I wanted to use the same prompts to create a portfolio site using Loveable. I wanted to see how far I could get with just the free version and also how well it would do in creating my site with just a few prompts. Well… it did well, but after 3 prompts, I ran out of credit! Here is how Lovable fared compared to Replit.

I started off by uploading my resume (in PDF format) and asking it to create a portfolio website for me (using the same prompt I used in Replit).

(caption - first prompt to create my portfolio site with resume attached)

It did not take that long for it to generate the site, but it ran into problems when generating the Preview. It gave me an error which I tried to dismiss multiple times.

I wanted to keep going. So I dismissed the error after trying to run it a couple of times.

Next, I went ahead and asked it to adjust the color and them and make the portfolio emphasize my AI experience. However, I forgot to attach my portfolio photo.

And Voila! It came up with a site with a very similar theme and design to the one by Replit.

However, it went ahead and generated a portfolio image to use. What was interesting is that it presented a photo similar to me (skin color and hair texture), which is very surprising and a bit interesting. Was that based on AI guessing what I would look like based on my name (origin, etc.) or based on my online presence? Inquiring minds want to know!!!

Because I wanted it to use my own photo, I went ahead and attached a professional photo of me and asked it to update the site.

Easy, peasy!

What I wanted to do next we tweak some of the content to make sure that it matched my work and experiences very closely - no hallucinations of stuff I never did!

But, sadly, I ran out of credit to do any more changes. Wonk wonk wonk 🙁

In the end, what was produced by Lovable was just as good as the one by Replit and impressive. However, I feel it would need a few more tweaks to be ready to be published. But, I am not quite ready to upgrade to the paid version. I will definitely come back to it after reviewing a few more tools.

Here are some of the pages it created. Unlike Replit, Lovable created just one single page (Replit had individual pages for each section)

If you are wondering how these tools work, I encourage you to give them a try. What they can do with just a few prompts, is nothing short of amazing.

I am now off to review a few AI tools for creating presentation slides. A few have already come in handy for my most recent projects. Stay tuned…

Replit claims it can turn an idea into a working site in minutes. So I asked: Could I build a personal portfolio using nothing but my resume?

I started off by uploading my resume (in PDF format) and asking it to create a portfolio website for me.

It took it about 7 minutes to generate all the files and sections to create the basic site.

Because I wanted something more than just a generic site with standard colors, I asked it to customize the color theme and include a professional photo of me that I attached to the prompt.

What it produced was very impressive and with just a couple more tweaks on the color effect, I was ready to finish and publish my portfolio site.

By the end, I had a working site in under an hour. It wasn’t perfect, but it erased starting-line friction.

Unfortunately, I was abruptly stopped from doing any more changes as I had run out of my free allotment. How disappointing! Nonetheless, what it produced with just a few prompts is nothing short of impressive. The site had several pages, with built in navigation and iconography that would take a long time to create without such a tool.

Here are some of the pages it created

What Replit was able to create was nothing short of impressive. With about 20-25 minutes, I was able to create what would have taken me days or months to accomplish. And if I had used the paid version, I am sure I could have gotten a lot further than I did. However, I must also note that there were parts that did not work as well for me. I wanted to make some more tweaks to the look and feel, but without knowing how to go in and modify the code manually, I could not get what I wanted done via the conversational UI. Again, it is possible I could have made more attempts to get it to work. I simply ran out of credit to keep trying. That said, Replit is still an incredible tool and will save many hours of work to develop and publish what it did in a few shots at describing the desired outcome. I would highly recommend giving it a go.

And now I’m off to try on another tool - Lovable. How will Replit compare the AI app builder with a $10B recent funding? Stay tuned for my review.