Full-stack IoT system with ESP32, FastAPI, React, and real-time MQTT
Delivered a full-stack smart home platform that integrates ESP32-S3 room-node firmware, MQTT messaging, a FastAPI backend, and a React dashboard for centralized monitoring and control of environmental and access workflows. The final system emphasized reliable subsystem integration, transparent validation evidence, and demo-ready operation across both live and simulated conditions.
Built an end-to-end IoT architecture spanning embedded firmware, backend APIs, messaging infrastructure, and frontend UX:
• Firmware/Hardware: Implemented and validated ESP32-S3 node workflows for environmental sensing and access-control behaviors, including staged bring-up and combined-load compatibility testing.
• Backend: Developed FastAPI services for ingest/control paths, health checks, websocket/broker interactions, and rules-oriented behavior.
• Frontend: Delivered a React dashboard with improved metric legibility and clear separation between Demo Mode and Live Mode.
• Infrastructure: Orchestrated services through Docker for repeatable local integration across API, broker, and supporting dependencies.
• Validation: Executed extensive backend pytest coverage and system-level timing/behavior experiments to support performance and reliability claims.
Case Study
Problem
Create a smart-home system that provides real-time visibility and control over environmental and security-related flows from one web interface, while remaining practical to validate and demonstrate under capstone constraints.
Architecture
- ESP32-S3 nodes -> MQTT broker -> FastAPI backend -> React dashboard
- Containerized services for repeatable local integration across API, broker, and dependencies
- Requirement-linked test and validation evidence to support final claims
Challenges
- Balancing hardware bring-up risk with software integration progress under tight final-deliverable timelines
- Managing firmware/hardware uncertainty while preserving reliable demo narratives
- Avoiding overclaims by separating measured end-to-end behaviors from partially exercised paths
Tradeoffs
- Prioritized staged integration and risk reduction over aggressive feature expansion
- Used Demo/Live workflow separation to improve reliability and communication clarity
- Chose evidence transparency over polished but unverifiable claims in final documentation
Outcome
Completed a multi-layer smart-home platform with validated backend reliability, functioning firmware-to-dashboard integration paths, and presentation-ready demo workflows. The project finished with strong software/test maturity and meaningful hardware integration progress, while clearly documenting remaining constraints and scope boundaries.
What I Learned
- How to deliver a credible embedded + cloud/web system by tying claims to measurable evidence
- Practical ESP32-S3 bring-up/debug strategy for real wiring, power, and integration constraints
- How MQTT, APIs, and containerized services behave under real integration pressure
- The value of explicit "fully demonstrated vs partially validated" reporting in engineering communication