Project Info

Solace

Devpost

This project did not submit a demo video on Devpost.

Inspiration

Social workers are the unsung heroes of our society—balancing large caseloads, bureaucratic hurdles, and emotionally draining tasks, often without the tools to match their mission. We wanted to build something that doesn’t just “use AI,” but genuinely serves the people who serve others. We were inspired by real interviews and research that revealed how overburdened systems leave both social workers and their clients underserved. SOLACE was born to fill that gap with empathy-driven AI tools.

What it does

SOLACE is an AI-powered assistant designed to support social workers with their most time-consuming responsibilities. Our platform allows social workers to: Log and manage important tasks (to-dos, deadlines, appointments) Auto-summarize long case notes using state-of-the-art AI Surface key client needs and connect them to local resources Work faster, with more clarity—without compromising on care

How we built it

Frontend: React + Tailwind CSS for a clean, mobile-responsive UI Mobile:React Native, Expo SDK 51+, React Navigation -Web: Next.js 15, React 18+, TypeScript, Tailwind CSS, Vercel deployment Backend: Node.js for routing and Python for AI processing AI Integration: Hugging Face BART model for classifying client case notes Sentence transformers for contextual resource matching Claude AI for monthly and quarterly report insight generation Data Design: Local-only sensitive data handling to preserve privacy

Challenges we ran into

Privacy vs Personalization: Designing a system that respects confidentiality while remaining context-aware Multiple AI Pipelines: Combining different models (NLP, embeddings) into a single cohesive product Time Constraints: Training useful models and implementing them meaningfully under strict hackathon time limits

Accomplishments we're proud of

Successfully integrating AI APIs and models for the first time as a team Implementing real-time voice input and processing tools Creating a product that doesn’t just look polished—but serves a purpose

What we learned

How to fine-tune and use pretrained NLP models for real-world tasks Key frontend skills in React and Tailwind Voice AI tools and how to work with APIs under pressure Collaborative development and ethical AI thinking, all under 36 hours

What's next

Expand to include appointment scheduling, live calling, and multi-user collaboration Build a client-facing version of the app so individuals can find help directly Improve AI accuracy with more training data, and add support for multilingual and mobile accessibility Launch a beta program with real social workers and gather feedback

Analysis

Compare with all teams

View

Metric

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only
  • AnthropicIn code
  • CSSIn code
  • FastAPIIn code
  • Hugging FaceIn code
  • JavaScriptIn code
  • Next.jsIn code
  • OpenAIIn code
  • PythonIn code
  • PyTorchIn code
  • ReactIn code
  • RedisIn code
  • SQLIn code
  • SupabaseIn code
  • Tailwind CSSIn code
  • VercelClaimed

14 of 15 appear in the indexed code. 1 claimed on Devpost could not be matched to code, which may simply mean the tool leaves no trace in the repository.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

669 KB

Source files

86

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.

0 stars