The Purpose

The LA Valley's Sepulveda Basin is fragmented and inequitable. Families are forced to walk hours between zones or drive, visitors are met with eroding infrastructure, and the only public transit here, the G-Line, has a $368 million 3-year closure, leaving people stranded now.

Our project envisions a comprehensive mobility hub system that integrates mobility, community, and sustainability, connecting all 2,000 acres. We designed to address the needs of the community, not just visitors, beyond 2028.

‍

‍

The Research

A site visit revealed unsafe, eroding infrastructure, and missing signage. We interviewed 5 visitors (3 locals, 2 new) who described the stress and barriers regarding community, mobility, and the climate. Forums echoed frustration about poor maintenance and lack of follow-through on changes, furthering distrust and sentiment of misused funds.

Key Findings

• Heat and lack of shade are decisive barriers
• ADA and low-income users are excluded from transportation equity
• Zones feel isolated with no wayfinding, many locals didn't even know about other zones
• Visitors prioritize safety and reliable connections when forming a community

‍

‍

The Approach

Ideation and Early Process

I facilitated the early design phase to give the team structure and direction. Our goal was to identify where intervention would matter most and create a clear workflow to move from raw ideas into testable concepts.

1. Framing the project — Led an ideation workshop to surface challenges and opportunities across the park.
2. Defining pillars — Mapped issues into three core areas: Community, Mobility, and Environment.
3. Assigning ownership — Each teammate took responsibility for one pillar, while I led Mobility and overall creative direction.
4. Early exploration — Guided the team through three weeks of sketching, feedback sessions, and refinement cycles.
5. Concept groundwork — Established the foundation for moving into higher-fidelity visualization and final design.

‍

Concept Finalization and Refinement

My focus was the mobility system and canopy design. I developed a biomimetic canopy that applied thermoauxetic metamaterials as a conceptual innovation. Auxetic weaves expand laterally when stretched and contract when compressed, giving them unique self-healing and load-resistant qualities. I combined this with negative thermal expansion (NTE) treatments so the canopy could dynamically respond to climate.

• In direct sun the weave tightens, closing gaps to create shade where and when it is needed most.
• As temperatures cool, the weave relaxes and reopens, allowing natural airflow and daylight.

‍

‍

This system proposed a low-maintenance, climate-responsive roof structure that adapted over time. The canopy integrates shaded pedestrian walkways, micro-mobility lanes, and a dedicated tram system that connects zones across the park.

‍

‍

Materials were selected to be resilient and ecological, including upcycled rubble from construction for walkways, fire-resistant wood, and native vegetation to embed biodiversity directly into the structure.

‍

‍

AI Workflow & Visualization Process

I built an iterative workflow to take sketches into high fidelity. This was a non-linear loop of sketching, editing, prompting, and refining until the concept was what I envisioned.

Hand sketch → refine prompts with ChatGPT → generate batches in MidJourney/Krea → select strongest image → edit in Photoshop → upscale with Topaz Labs → if perspective was missing, model quickly in Spline → screenshot → feed back into MidJourney as reference → repeat

For perspectives that were impossible to generate, I built rough 3D forms in Spline, screenshotted the view, and used that as an omni-reference to guide AI generation.

‍

‍

I applied the same loop to generating videos with MidJourney, guided by starting and ending frames. A/B testing showed that prompts expanded with ChatGPT consistently produced stronger results than my base prompts. Using this workflow I produced still renders, material boards, interiors, and a full one-minute cinematic vision video.

‍

‍

The Results

The concept reimagined the Basin as an adaptive ecological system. The thermoauxetic canopy would cut heat stress, support biodiversity, and sustain LA’s climate.

I taught my team and other groups how to use AI workflows to unlock and communicate ideas at a higher fidelity. This empowered everyone to visualize their concepts with clarity and consistency, not being curbed by their skillset.

BMW Designworks responded strongly to our concepts. They were fascinated by the thermoauxetic material innovation and impressed by how speculative research was paired with real systems design.

‍

Future opportunities

• Revenue model — potential business applications tied to canopy and material systems
• AR partnerships — collaboration with companies like Snapchat to create unique immersive canopy experiences
• Community integration — involving students and local artists in AR murals and kiosks, turning hubs into cultural anchors

‍

These results solidified our solution as a vision for a multi-modal, community-driven, and climate-positive hub that goes beyond 2028 event needs to serve LA communities.

‍

The Purpose

The LA Valley’s Sepulveda Basin is 2,000 acres of parkland with some of the region’s most diverse amenities. It has Lake Balboa, golf courses, sports fields, and a wildlife preserve. In reality, the Basin is fragmented and inequitable. Families walk for hours between zones, visitors encounter eroded paths and failing infrastructure, and the only public transit option, the G-Line, is shut down for three years at a cost of $368 million, leaving people stranded.

The city’s Vision Plan centers ecology while ignoring mobility. The result is a park that looks complete on paper but functions as disconnected zones. Our project reframed the Basin as a comprehensive mobility system that integrates community, movement, and sustainability across the full 2,000 acres. The design was intended for residents first, while still preparing the park as a global stage for 2028.

‍

‍

The Research

We started with lived experience. A site visit exposed eroded infrastructure, broken circulation paths, and missing signage. My group interviewed five visitors (three locals, two new) who each described different forms of stress: unshaded distances, unsafe and unmarked paths, and visitors unaware entire zones even existed.

To capture broader sentiment, we then reviewed forums, Yelp reviews, Reddit discussions, and council notes. The same complaints surfaced repeatedly: unsafe trails, trash and encampments, lack of lighting, and an ongoing distrust of the city’s promises. People spoke openly about misused funds and skepticism that anything would change.

Key Findings

• Heat and lack of shade are decisive barriers
• ADA and low-income users are excluded from transportation equity
• Zones feel isolated, many locals did not know about other areas
• Safety, shade, and reliable connections define how community forms

‍

‍

The Approach

Ideation and Early Process

I designed and led an ideation workshop to bring structure to the ideation phase of our project. We mapped pain points across the park and distilled them into three design pillars: Community, Mobility, and Environment. Each teammate took ownership of one pillar. I led Mobility and held creative direction for the project. This structure gave the team freedom to explore while ensuring all ideas aligned into a cohesive system.

We spent three weeks sketching, testing, and refining. I created clear feedback loops by facilitating team co-working and during studio sessions so concepts evolved quickly, and positioned our process to move from low-fidelity sketches into system-level design quicklly without compromising quality.

‍

Concept Finalization and Refinement

My focus was the mobility system and canopy design. From the beginning, I wanted a structure that wasn’t ornamental but truly embedded into the ecology and climate of the Valley. The canopy had to solve multiple problems at once: extreme heat, lack of wayfinding, disconnection between zones, and the need for safe multi-modal movement.

I designed a biomimetic canopy that applied biotechnology through thermoauxetic metamaterials. Auxetic structures behave in the opposite way of most materials: when stretched, they expand laterally instead of thinning, which makes them resilient under pressure, resistant to tearing, and capable of self-healing behavior. This property makes them ideal for environments where tension, stress, and shifting climate are constants.

‍

I paired this with a conceptual application of negative thermal expansion (NTE) compounds. The idea was to create a temperature-responsive weave that tightens in extreme heat and relaxes as conditions cool.

• In the Valley’s harsh summer sun, the weave would contract, closing its gaps to provide denser shade and direct heat protection exactly where needed.
• In cooler conditions, the weave would relax and reopen, allowing airflow and natural daylight to filter back in.

This created a passive, climate-adaptive roofing system — one that requires little to no maintenance because the material itself performs the environmental response.

‍

‍

Beyond shading, the canopy was designed as the anchor for mobility and wayfinding. Each hub included shaded pedestrian lanes for safety and comfort, micro-mobility lanes for bikes and scooters, and a dedicated tram connector to unify zones across the 2,000-acre park. The form worked from micro to macro, scaling from individual comfort to system-wide connectivity.

Materiality reinforced the design intent:

• Upcycled rubble from construction was reused for walkways and mobility lanes
• Fire-resistant wood ensured longevity in wildfire-prone areas
• Native vegetation was cultivated into the structure itself, reinforcing local ecosystems while providing cooling and biodiversity
• The auxetic NTE-treated weave formed the conceptual innovation, opening a pathway for new material research

‍

‍

AI Workflow & Visualization Process

I built the AI workflow that powered both our design fidelity and our ability to communicate the vision. The process was iterative and circular, blending hand sketches with AI generation and manual refinement until I achieved precision.

Hand sketch → refine prompts with ChatGPT → generate batches in MidJourney/Krea → select strongest image → edit in Photoshop → upscale with Topaz Labs → if perspective missing, model quickly in Spline → screenshot → feed into MidJourney as reference → repeat

When perspectives were impossible to generate, I used Spline to model rough geometry, captured the angle, and reintroduced that into MidJourney as a reference. This hybrid method gave me control over form and perspective without losing the speed of generative workflows.

‍

‍

I extended this method into motion, scripting and producing our one-minute vision video. Using starting and ending frames as anchors, I generated sequences in MidJourney. A/B testing showed that prompts expanded through ChatGPT consistently delivered stronger results than my base prompts. This process produced consistent imagery, motion, and atmosphere across the entire project. As AI Lead, I trained teammates to apply this workflow themselves. This raised the quality of everyone’s output and allowed the team to communicate their ideas at a fidelity that matched their vision.

‍

‍

The Results

The concept reframed the Basin as an adaptive ecological system. The thermoauxetic canopy reduced heat stress, supported biodiversity, and introduced sustainable construction methods that directly addressed the Valley’s climate.

My leadership extended beyond design delivery. By teaching and empowering my team and neighboring groups to apply AI in their creative process, I gave them tools to visualize and communicate at a professional level regardless of sketching and modeling ability. This amplified creativity across the internship and positioned our team as a leader in speculative prototyping.

BMW Designworks responded with strong interest. They were fascinated by the thermoauxetic material concept and impressed by how speculative material exploration had been paired with a grounded mobility system. Feedback highlighted potential business opportunities, including revenue models tied to canopy applications.

We also began exploring external partnerships. Collaborating with AR companies like Snapchat could add immersive experiences to each canopy. These installations could connect to kiosks and wayfinding, creating unique attractions across the Basin. Involving local students and artists to produce AR murals and experiences would make the hubs into cultural anchors, blending transit, art, and community development.

Future opportunities

• Revenue model — potential business applications tied to canopy and material systems
• AR partnerships — collaboration with companies like Snapchat to create unique immersive canopy experiences
• Community integration — involving students and local artists in AR murals and kiosks, turning hubs into cultural anchors

‍

The project evolved into more than mobility planning. It became a multi-modal, community-driven, and climate-positive system designed to serve Los Angeles residents while also functioning as a global showcase of equity and innovation well beyond 2028.

‍

‍