I'm Rajjit Laishram, an IoT Software Developer and Autonomous Systems Engineer based in Imphal, Manipur. I specialize in building the software layer between physical hardware and intelligent decision systems β from embedded edge devices to AI-driven monitoring platforms and autonomous drone systems.
Currently at Nibiaa, building enterprise IoT infrastructure with hybrid LoRaWAN/Satellite monitoring, offline AI anomaly detection, and MCP-based tool orchestration. Previously led autonomous drone software development at NAWA, reaching the finals of India's NIDAR Disaster Management competition with 215+ real-world flight tests.
"I write the code that makes machines think."
IoT Software Developer β Nibiaa (July 2025 β Present)
- Designed and demonstrated a prototype hybrid indoor/outdoor asset tracking system using LoRaWAN and satellite communication β built a simulated monitoring dashboard to evaluate the use case for an enterprise equipment rental client.
- Built a custom AI alerting application with geofencing, restricted zone breach detection, and AI-driven anomaly recommendations integrated directly into the IoT platform.
- Researched and developed MCP connectivity with Ollama for offline AI inference inside IoT platforms β keeping all device data fully on-premise with zero cloud dependency for AI processing.
- Customized a proprietary IoT mobile application for client use, enabling on-the-go device monitoring and control without requiring access to a full web dashboard.
- Engineered a peer-to-peer mesh communication layer for proprietary IoT network use.
- Deployed unified system monitoring integrating analytics, CRM, and custom APIs for platform observability.
- Mentored an intern on AI integration patterns and MCP-based tool orchestration within IoT systems.
- Built a complete autonomous drone software stack from scratch β custom Electron-based desktop GCS, YOLO + Hailo AI survivor detection pipeline (~80% accuracy), switchable RTSP/MJPEG video streaming, and multi-drone coordination logic.
- Implemented autonomous mission execution using DroneKit and MAVLink with multi-GPS cross-validation and custom failsafe routines.
- Reached the finals of India's NIDAR Disaster Management competition. Validated across 215+ real-world flight tests.
- Developed mobile drone control interfaces and engineered mission control algorithms for autonomous flight path execution in disaster simulation environments.
- Hardware: Raspberry Pi, ESP32, Arduino, NodeMCU, Pixhawk
- Protocols: MQTT, CoAP, WebSockets, RTSP, LoRaWAN, I2C, SPI, UART
- Edge & Connectivity: Edge Computing, Embedded Linux, DroneKit, Satellite Communication
- Flight: DroneKit, MAVLink, Multi-GPS Cross-Validation, Failsafe Logic
- Computer Vision: OpenCV, YOLO, Hailo AI Architecture
- Streaming: RTSP, MJPEG, Custom Video Stream Switching
- Offline AI: Ollama, MCP Tool Orchestration
- Frameworks: FastAPI, Node.js, Electron
- Cloud: Azure IoT, GCP, Docker
- Frontend: React, Next.js, TypeScript, Tailwind CSS
- Mobile: React Native
- Databases: PostgreSQL, MongoDB, Appwrite
Complete autonomous drone software stack built from scratch. Custom Electron desktop GCS + YOLO/Hailo AI survivor detection (~80% accuracy) + switchable RTSP/MJPEG streaming + multi-drone coordination. Validated across 215+ real-world flight tests. NIDAR Finalist.
Stack: Python Β· FastAPI Β· DroneKit Β· MAVLink Β· OpenCV Β· YOLO Β· Hailo AI Β· React Β· Tailwind CSS Β· Electron Β· PostgreSQL Β· WebSockets Β· Raspberry Pi Β· Pixhawk
πΈ Field Documentation & Competition Posts
Web-based virtual laboratory enabling students to conduct chemistry and physics experiments without physical equipment. Visit Lab
Stack: Next.js Β· Appwrite Β· Tailwind CSS
Privacy-first ephemeral chat application. Conversations auto-delete β zero permanent data traces. Visit App
Stack: Next.js Β· TypeScript Β· Tailwind CSS
Web-based music player with audio upload, streaming, and visual waveform playback. Visit Site
Stack: React Β· Appwrite Β· Tailwind CSS
A personal project showcase platform with a public-facing gallery backed by Google Sheets, a secured admin panel for content management, and a terminal-inspired visual design. Visit Projects
Stack: React Β· Tailwind CSS Β· Google Sheets API
Custom-designed and built from scratch β an educational platform offering tech knowledge through blogs, downloadable notes, interactive quizzes, and a wallpaper gallery. No templates, no website builders. Visit Blog
Stack: React Β· Tailwind CSS Β· React Router Β· Appwrite
Last updated: May 2026