Inspiration Leveraged 13 years in private banking with HNW clients (Ben) to spot the need for unbiased, holistic advice—driven by misaligned incentives in proprietary and commoditized products—and recognized that AI, open banking, and modern financial infrastructure make now the ideal time to build a solution.
What Is It Launched a savings-account pilot: three client profiles fed into an AI engine that recommends whether to move funds among five major banks.
Tooling Built on a versatile stack—Supabase for data management, Python for scripting, while integrating OpenAI & Perplexity for AI, LangChain & Groq for model workflows, Tavily for web search, Proto.io and MagicPattern for rapid prototyping and design, Gamma for presentation, FDIC & Investing.com for CDS data, plus Google Sheets/Docs and GitHub for collaboration.
Challenges Navigated tight timelines and asynchronous collaboration across three newly combined experts, rapidly selecting and deploying flexible tools.
Point-of-Pride Demystified complex financial instruments (e.g., credit default swaps) into accessible models for non-engineers, showcasing the value of client-, industry-specific, and public data.
Learning Engineered a heavy-lift data pipeline and trained a compact AI model to deliver personalized expertise at scale—data needs will grow as Quin tackles more finance categories.
Next: recruit engineers and secure funding for a five-month pilot to expand client work, refine model training, and drive toward revenue.
Log in or sign up for Devpost to join the conversation.