My Role
Multimodal AI product engineer
All Projects
Interactive prototypeFlagship case study
Fixr
AI-Powered Real-Time Tech Support Assistant
Remote technical support loses the visual and conversational context a technician uses to diagnose a device safely and efficiently.
Interactive preview
Full demoWhat I Built
AI-powered troubleshooting tool using text, images, and live camera input to diagnose hardware issues like loose cables, damaged components, overheating, and blinking error patterns, then delivers step-by-step repair guidance instantly.
Evidence
Working interface, documented system behavior, and implementation-level decisions.
Technical Architecture
From system input to explainable output.
The control gate is shown as a first-class stage, not an afterthought added around the workflow.
Technical Decisions
- Built a multimodal troubleshooting assistant that accepts text, still images, and live camera input, so a user can point a phone at the problem instead of describing it.
- Streamed responses so guidance appears progressively instead of after a long silent wait.
Security Considerations
- The public case study makes no additional security claim beyond the documented controls shown in the architecture.
Outcome & Evidence
- Mapped detected symptoms to ordered repair steps, with the safety-relevant steps such as power isolation surfaced before anything else.
Google AI StudioComputer VisionReal-timeMultimodal AI
Working product
Try the interactive demo.
The product experience is part of this case study. Explore it here, reset its state, or switch viewport sizes without leaving the project page.
fixr.zainkhalilkhan.com
Fixr
AI hardware support
FixrWorkspace5 updates
FixrAUTOMATED HARDWARE TECHNICIAN
OnlineHardware assistant
Describe the issue and Fixr will narrow it down
Power railOverheatingLoose cableLED patternComponent
Fixr can make mistakes. Verify repair steps before opening hardware.
Zain Khalil Khan
Client-side sandbox. State is in memory and nothing is sent to a server.
Next Case Study