Everything in one place — the challenge, the two-week build, the Coda workspace, the ethics, the awards, and the day itself.
The PMI-LA × Sound of Earth Innovation Challenge brings together students, project managers, sustainability professionals, technologists, nonprofit partners, community leaders, and emerging builders to create practical tools, apps, and resources that help Los Angeles communities and advance sustainability.
AI has made it dramatically easier to build a prototype — so the real test is everything around the build: identifying a problem worth solving, planning the work, collaborating, using AI responsibly, and delivering something people can actually use. That's project management, and it's what this challenge is built to develop.
Ship a working prototype — a tool, app, dashboard, or resource hub — that demonstrates a viable approach to a real community problem. Any platform, any stack.
Practice real project management in an AI-enabled environment: scope, stakeholders, risk, collaboration, ethical tech use, and outcome-based delivery — captured in Coda.
Solutions are designed for and with the communities they serve, grounded in sustainability, equity, accessibility, and climate resilience.
More than a competition — a learning experience, a project-management exercise, and a community-impact initiative, built in community, for Los Angeles.
Every team works from a real problem brought by a nonprofit or community partner — a genuine pain point, with a real population, real constraints, and a real definition of what "better" looks like. Pick the brief that moves your team and build a practical tool, app, or resource for it, guided by the principles below.
Browse the real problem briefs submitted by our nonprofit and community partners and pick the one your team wants to take on.
Design and ship a tool, app, dashboard, or resource that addresses it directly — grounded in that organization's real needs and constraints.
Every solution should reflect alignment with these five principles.
Consider environmental impact, resource efficiency, or climate resilience.
Designed for and with the communities the solution aims to serve.
Usable by and beneficial to underserved or frontline populations.
Consider data privacy, algorithmic fairness, and responsible AI use — see Responsible AI & Ethics.
Demonstrate intentional planning, execution, and reflection.
Wildfire resilience, climate preparedness, community resource navigation, solar access for low-income households, circularity, and sustainability education — inspiration, not requirements.
From framing a real problem, to building and documenting, to presenting live — the challenge mirrors how good work actually ships. Two required workshops set the foundation before the build begins.
Before the build begins, nonprofit and community partners present real pain points through a standardized problem brief covering the population they serve, the specific workflow or resource-access problem, what "better" looks like, and what teams should plan for. Each team picks the brief that resonates and builds directly for it — every project is grounded in a real organization's needs.
Teams design and deliver two things in parallel:
Teams present in person to judges. Each presentation includes a live demo, a walkthrough of how the team planned, scoped, divided work, managed risk, iterated, and delivered, and an honest reflection on what worked and what they'd do differently.
Two live workshops run in the kickoff window — both required before teams start building. They're recorded so every participant can attend regardless of schedule.
Every participant completes this workshop. It covers responsible and ethical AI use for community-facing tools: data privacy and consent, bias and fairness, transparency and disclosure, human oversight, the environmental footprint of AI compute, and designing with communities rather than for them. Sets the standard the whole challenge is held to.
A deeper technical session for the PM side: planning, scheduling, scope and stakeholder management, risk, and earned-value fundamentals — applied to a fast, AI-enabled sprint — plus a CSPP review. confirm CSPP scope Grounds the project-management package teams build in Coda.
Target dates — subject to confirmation.
Every team turns in two deliverables: a functional prototype, and a structured PM package hosted in Coda. The Coda workspace is a living record during the build — and a professional artifact afterward that showcases end-to-end project delivery.
Each team receives a copy of a single Coda doc with embedded guidance on each page — what the artifact is, why it matters, and what good looks like. It's self-serve: structure and examples are provided, but how you apply PM principles to your project is up to you. That's part of what's evaluated. Customize, extend, or restructure it freely — it's a starting point, not a constraint.
A tool, app, resource hub, dashboard, or other digital product that demonstrates a viable approach to your chosen problem — built in any platform or stack your team prefers.
Your completed Coda workspace: charter, stakeholder analysis, task breakdown and tracking, risk register, a sustainability plan, and retrospective — each mapped to a standard PMI / PMBOK process area, and together the documented story of how you delivered.
Submit a link to your prototype and your Coda workspace through the official submission form by the deadline — Oct 3, 2026 (target). Judges review submissions offline ahead of the in-person event, where teams present a live demo and a walkthrough of their process. Your Coda workspace must include the required AI Use Disclosure. All submissions are open-source and non-commercial — built for the communities they serve, not for private profit. Full format and file-type details land in the Coda resource hub. license: CC BY-NC / AGPL TBA
Required for all participants The AI Ethics workshop, led by Sound of Earth, is mandatory before building begins.
When anyone can generate a working prototype in an afternoon, the hard part is no longer can you build it — it's should you, for whom, and at what cost. These tools are aimed at frontline and underserved communities, so a mistake isn't a bug report; it's a real person not getting a resource they needed. Responsible AI use is not a compliance checkbox here. It's the work.
Every team commits to the principles below, and every participant completes the mandatory AI Ethics workshop led by Sound of Earth before building begins. We're also clear about scope: challenge submissions are prototypes, not production systems — they shouldn't be put in front of real community members without further testing, consent, and partner sign-off.
It often is. Data centers are straining power, water, and air in communities across the country — frontline communities most of all, and we won't pretend otherwise. But we also believe AI isn't going away; it's already woven into how the people who've adopted it live and work. So we take that as a starting fact and turn it into the challenge's hardest constraint: build as if truly sustainable AI is the goal. It doesn't fully exist yet, and this challenge is meant to be one small step on the path toward it.
Sound of Earth has team members researching data-center design, the policy shifts shaping where and how they get built, and the ethical impact on the frontline communities living beside them. That work sits at the center of this challenge, not at its edge.
Don't put real community members' personal data into AI tools without informed consent. Prefer synthetic, aggregated, or public data. Collect the minimum you need, and say what you collect.
AI models inherit the biases in their training data. Test how your tool behaves across the full range of people it serves — especially those at the margins — and never automate a decision that could quietly harm them.
Be honest about where AI is used — in your product and in your Coda package. Label AI-generated content, cite your sources, and don't present a model's output as verified fact without checking it.
Keep a person in the loop on anything consequential. The model is a tool; your team is accountable for what it produces. Design for a human to catch, correct, and override.
AI is not free, environmentally. Training and inference carry real energy and water costs. In a sustainability challenge, ask whether AI is the right tool for the job — and when it is, use it efficiently.
Beware techno-solutionism — not every problem needs an app. The strongest solutions are shaped by the community they serve and are accessible by default (plain language, assistive-tech support, low-bandwidth, multiple languages where it matters).
Each team includes a short AI Use Disclosure in its Coda workspace: which AI tools you used, where you used them, what you verified, and what you chose not to automate. It's good project-management and risk practice — and it makes the Best Use of AI award about judgment, not just output.
"The goal isn't to use the most AI. It's to use it well — transparently, accountably, and in service of the people on the other side of the screen."
Challenge ethics commitment
Awards are category-based and tied to partner missions, so a team that tackled a completely different problem can still win a nonprofit's award if the work deeply advances their mission. Categories to be finalized with partners.
Prizes are funded by sponsors and partners — expect a mix of cash, sponsor credits, recognition, and connections to mentors and recruiters. Specific amounts are confirmed closer to the event. amounts TBA
Each sponsoring nonprofit names an award aligned with its mission and selects the team whose solution best advances that mission area — regardless of which problem the team chose. One per sponsoring partner. Example categories:
Categories finalized with partners. in progress
Cross-cutting awards celebrating standout craft and approach:
Guidance is primarily self-serve. Teams are expected to self-organize and apply PM principles independently — the ability to plan and execute with the resources at hand is part of what the challenge evaluates.
The pre-built Team Workspace with embedded instructions on each PM artifact.
A session walking through PM methodology expectations and the Coda workspace. link TBA
Written examples of what strong project artifacts look like. examples TBA
Standardized partner briefs to choose from as reference material. briefs TBA
Encouragement to use modern AI tools — Claude, Coda, code-generation tools — with responsible-use guidance.
Lightweight project-management guidance drawn from PMI practice, scaled for a fast, AI-enabled innovation sprint.
Background reading, datasets, and frameworks on climate resilience, equity, accessibility, and community impact. links TBA
Challenge FAQs and submission instructions via the Coda resource hub. hub TBA
PMI-LA member mentorship is being explored as an additional layer of support — the facilitation model is still being defined. Regardless, teams are expected to self-organize and apply PM principles on their own.
Anyone 18 or older. You don't need to be a student, a PMI member, or a developer — students, project managers, technologists, designers, and community builders are all welcome, and mixed teams of students and working professionals are encouraged. You don't have to live in Los Angeles, but the challenge focuses on LA communities and the final event is in person in LA. confirm at launch
No — registration is free for participants. (Nonprofit and community partners take part for free too.)
Teams are typically 2–5 people. You can register an existing team or sign up solo and be matched into one at the kickoff call. size to confirm
Yes — PMI members earn 10 PDUs for taking part, on top of a portfolio-ready project artifact and PMI-LA community access. PDU category to confirm
You choose from the real problem briefs submitted by our nonprofit and community partners, then build directly for the one your team picks. Every project is grounded in a real organization's needs — that's the heart of the challenge.
Any platform or stack — web app, Coda doc, Figma prototype, mobile app, and more. The project-management package, however, lives in the provided Coda workspace.
Yes — you're encouraged to use modern AI tools like Claude, Coda, and code-generation tools responsibly. Using AI well is part of what's being evaluated.
A functional prototype and a completed Coda project-management package. On event day, teams present a live demo plus a walkthrough of their process.
Experienced project managers, technologists, sustainability leaders, and nonprofit partners evaluate submissions across both product and process.
Judge & mentor lineup coming soon
Target date: Saturday, October 17, 2026 · 8:30 am – 4:30 pm alt: Oct 10 · venue TBA
Every registered team that submits on time presents on event day — there's no cut to finalists. Bring your team, demo your work, and be recognized.
Arrival, registration desk, and team check-in to start the day.
Each team demonstrates its working prototype to the judging panel.
Teams present how they planned, scoped, managed risk, iterated, and delivered.
An honest look at what worked, what didn't, and what each team would do differently.
Time to connect with mentors, judges, sponsors, partners, and fellow participants.
Overall placements, nonprofit mission awards, and special recognition to close the day.
Free to enter. Register as a team of 2–5, or sign up solo and we'll match you into a team at kickoff. team size to confirm Sponsors, mentors, judges, and nonprofit partners welcome too.
Add your name (and teammates, if you have them). It's free, and takes a couple of minutes.
We confirm you and share the Coda resource hub — problem briefs, the workspace template, FAQs, and key dates.
Bring your own team or get matched at the kickoff call, then start building.
Get updates on dates, registration, problem briefs, and resources as they're announced.