Embedded Systems Engineer — Edge AI & Sensors

Job Description

This role sits at the point where our hardware, our sensors and our AI models meet. A model that hits 92% mIoU in a notebook is worth nothing if the device it runs on drops frames, loses GPS lock, overheats in a Rajasthan summer or produces uncalibrated thermal data. Closing that gap is the job. You will work across our product lines — building the capture stack for dashcam units, standing up Jetson Orin nodes, and integrating thermal and LiDAR sensors. You will work directly with the AI team on what the models need from the hardware, and with the hardware manager on what the hardware can actually deliver. This is a hands-on role in a small team. You will spend time at a bench with a soldering iron and a logic analyser, time in a terminal cross-compiling and profiling inference, and time in the field on a highway shoulder or at a remote installation site figuring out why unit 14 stopped reporting.

Responsibilities

• Bring up and configure edge computing devices and embedded Linux systems.
• Set up NVIDIA Jetson and ARM platforms.
• Integrate RGB, thermal, LiDAR, depth, IMU, and GPS sensors.
• Calibrate sensors and synchronize multi-sensor data.
• Deploy, convert, and optimize AI models using ONNX and TensorRT.
• Build and maintain on-device software, inference pipelines, buffering, and GPS tagging.
• Manage cloud uploads, OTA updates, device telemetry, and field deployments.
• Debug field devices and manage power, thermal issues, wiring, calibration, and BOM documentation.

Required Skills:

• LiDAR
• SDK Development
• GMSL
• MIPI CSI
• V4L2
• NVM
• eMMC
• Fastboot
• U-Boot
• Device Tree
• BSP Development
• ARM SBC
• NVIDIA Jetson
• Embedded Linux
• Buildroot
• Yocto
• Model Quantization
• TensorRT
• ONNX
• ROS2
• ROS
• PTP
• Kalibr
• OpenCV
• NMEA Parsing
• UART
• SPI
• I2C
• GPS Interfacing
• IMU Interfacing
• Shell Scripting
• Git
• KiCad
• Oscilloscope
• Hardware Debugging
• IP Rating
• Network Resilience
• Power Budgeting
• Balena
• AWS IoT
• OTA Updates
• MQTT
• SQLite
• Multithreading
• Python
• DeepStream SDK
• C++

Qualifications

• 2+ years of hands-on experience in embedded Linux, edge AI deployment or sensor systems engineering. Strong candidates with 2 years of genuinely deep work will be preferred over 5 years of shallow exposure.
• Proficient in Python and C++ — enough C++ to work inside an inference pipeline, not just script around one.
• Comfortable on embedded Linux on ARM: JetPack or Yocto/Debian based images, systemd services, device tree basics, cross-compilation, kernel modules and driver troubleshooting.
• Working knowledge of hardware interfaces — I2C, SPI, UART, GPIO, USB3, MIPI CSI-2, Ethernet — and the ability to debug them with a multimeter, logic analyser or oscilloscope rather than by guessing.
• Practical experience with camera and video pipelines: V4L2, GStreamer, RTSP, encoding and frame timing.
• Experience deploying a trained neural network onto an edge device — ONNX Runtime, TensorRT or equivalent — including quantisation and measured latency benchmarking.
• Git, structured version control habits, and the discipline to document what you built.
• Willingness to travel for field deployments and to work outdoors at installation sites.

Pay: ₹500,000.00 - ₹800,000.00 per year

Work Location: In person