Sight2Sense

• 

  GUI Interface

Inspiration

My inspiration came from working with electronics and realizing that most failures aren’t caused by complex problems, but by missing one basic step — power, polarity, grounding, or understanding what a component actually does.

Beginners often feel stuck because there’s no one around to guide their thinking in real time. In a lab, an experienced mentor naturally asks the right questions and helps you reason through the problem. I wanted to recreate that experience using AI — an assistant that doesn’t just give answers, but helps users think like an engineer.

That idea became Sight2Sense

What it does

Sight2Sense is an AI-powered electronics assistant that helps users design, analyze, and troubleshoot electronic circuits, PCBs, and embedded systems.

Users can: Sight2Sense is an AI-powered electronics assistant that helps users design, analyze, and troubleshoot electronic circuits, PCBs, electronics and embedded systems.

-Ask electronics and embedded systems questions -Upload images of circuits or PCBs for analysis -Receive step-by-step troubleshooting guidance -Get safety warnings for high voltage, polarity, and battery risks -Learn concepts in a way that adapts to their skill level

Instead of giving generic answers, Sight2Sense explains why each step matters, just like a real mentor would in a lab. -Ask electronics and embedded systems questions -Upload images of circuits or PCBs for analysis -Receive step-by-step troubleshooting guidance -Get safety warnings for high voltage, polarity, and battery risks -Learn concepts in a way that adapts to their skill level -Instead of giving generic answers, Sight2Sense explains why each step matters, just like a real mentor would in a lab.

How I built it

I built Sight2Sense using Google Gemini via Google AI Studio, leveraging its strong reasoning and image understanding capabilities. The frontend was developed with Vite and TypeScript to ensure fast performance and a clean user experience.

The core of the project is a carefully designed system instruction that forces the AI to behave like an electronics mentor: -Structured responses -Safety-first reasoning -Clear diagnostics flow (power → logic → output) -Mobile-friendly explanations

Rather than relying on raw AI output, I focused heavily on prompt engineering and response control to make the assistant reliable and educational

Challenges I ran into

One of the biggest challenges was ensuring that the AI remained helpful without becoming unsafe or misleading, especially when dealing with real hardware and electrical risks.

I also faced challenges with: -API key handling and deployment decisions -Making explanations concise but still technically accurate -Designing responses that work well on mobile devices -Ensuring image-based analysis stayed practical and cautious

Each issue required careful iteration and testing.

Accomplishments that I am proud of

I’m most proud of building a system that feels genuinely useful rather than just impressive.

Sight2Sense:

-Responds like a real electronics mentor -Prioritizes safety in every interaction -Adapts explanations based on user skill level -Helps users learn how to think, not just what to do

Building this project alone, end-to-end, from concept to working AI system, was a major personal achievement.

What I learned

Through this project, I learned that building with AI is as much about design and responsibility as it is about technology.

I gained experience in: -AI prompt engineering -Electronics-focused reasoning workflows -Frontend development and integration -Responsible AI usage in real-world scenarios

Most importantly, I learned how to turn AI into a teaching tool instead of just an answer engine.

What's next for Sight2Sense

Next, I plan to expand Sight2Sense with:

-More interactive troubleshooting flows -Component-level identification improvements -Guided PCB debugging checklists -Offline reference support for common electronics tasks

My long-term goal is to make Sight2Sense a trusted learning companion for students, hobbyists, and engineers working with electronics.

Built With

• and
• and-deployment-google-gemini-api-?-multimodal-ai-reasoning-and-image-understanding-prompt-engineering-?-structured-system-instructions-for-reliable-electronics-mentoring-web-apis-?-image-upload
• input-handling
• maintainable-application-logic-vite-?-fast-frontend-build-tool-and-development-environment-html-&-css-?-clean
• responsive-user-interface-google-ai-studio-?-model-configuration
• testing
• typescript-?-for-type-safe
• ui