The latest articles on DEV Community by srishti (@srishti_2870a7693a3c01b49). https://dev.to/srishti_2870a7693a3c01b49 https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3645595%2Fcc0883d1-d702-42f6-89c1-43b3518be044.jpg https://dev.to/srishti_2870a7693a3c01b49 en srishti Thu, 04 Dec 2025 08:05:40 +0000 https://dev.to/srishti_2870a7693a3c01b49/finaura-your-financial-glow-up-assistant-4cjj https://dev.to/srishti_2870a7693a3c01b49/finaura-your-financial-glow-up-assistant-4cjj <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fqzx8k15fdw6ncdedij0g.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fqzx8k15fdw6ncdedij0g.png" alt=" " width="800" height="705"></a></p> <p>🚀 I signed up for the 5-Day AI Agents Intensive with Google and Kaggle not just to learn—but to build something that matters. As a creative educator and Python developer, I’ve always believed that technical tools should feel intuitive, empathetic, and empowering. That’s how Finaura AI was born—a financial agent designed to help users navigate data with clarity and emotional resonance.</p> <p>🧩 Building Finaura: From Idea to Agent<br> Tech Stack Highlights:</p> <ul> <li>Google ADK for agent orchestration</li> <li>Gemini 2.5 Flash for fast, context-aware generation</li> <li>Custom Python workflows for encoding, scaling, and regression modeling I architected Finaura to handle real-world financial data, but I didn’t stop at functionality. I wanted the agent to feel like a guide—not just a calculator. So I layered in storytelling: every insight comes with context.</li> </ul> <p>🔧 Debugging the Soul of the Agent<br> The biggest breakthroughs came from troubleshooting.</p> <ul> <li>I iterated hybrid encoding pipelines until they were robust enough for noisy datasets.</li> <li>I refined regression workflows to balance performance with interpretability.</li> <li>I learned how to teach through errors—turning bugs into lessons for my audience.</li> </ul> <p>🎭 Storytelling Meets Strategy<br> Finaura isn’t just a tool—it’s a character.<br> I crafted its voice to be calm, clear, and confidence-boosting.</p> <p>📚 What I Learned</p> <ul> <li>Agents are more than workflows—they’re experiences.</li> <li>Debugging is design. Every error is a chance to improve the user journey.</li> <li>Branding matters. A memorable name, a punchy tagline, and a clear value prop make your agent unforgettable.</li> <li>Community fuels growth. Sharing my build process helped me connect with other creators and learners.</li> </ul> <p>💬 Final Reflection<br> This challenge wasn’t just about building an agent—it was about building belief.<br> Belief that I could blend creativity with computation.<br> Belief that education can be emotional, technical, and transformative.<br> Belief that agents like Finaura can help people feel less lost in their data—and more in control.</p> googleaichallenge ai agents devchallenge