Skip to content
AESTECHNO
Updated  20 min read Hugues Orgitello EN

NVIDIA Jetson Orin: embedded AI from board to product

NVIDIA Jetson guide for embedded AI: Orin Nano, Orin NX, AGX Orin. GPU architecture, robotics and vision, custom carrier board by AESTECHNO Montpellier.

Choosing a Jetson starts with fixing a power budget: across the Orin family, performance sits around 5 TOPS per watt, from the Orin Nano (7-15 W) up to the AGX Orin (up to 60 W). NVIDIA Jetson processors are compact, energy-efficient compute modules that run Artificial Intelligence (AI) models directly in the field, with no cloud dependency. Whether the target is autonomous robotics, industrial vision, autonomous vehicles or intelligent surveillance, the Jetson family delivers inference performance that opens new ground for embedded systems.

Key takeaways

  • Jetson Orin lineup: Orin Nano (40 Total Operations per Second, TOPS, 7-15 W), Orin NX (70-100 TOPS, 10-25 W), AGX Orin (up to 275 TOPS, 2048 Compute Unified Device Architecture (CUDA) cores, 64 GB LPDDR5, 15-60 W).
  • Software stack: JetPack 6.0 on Ubuntu 22.04, CUDA 12.2, TensorRT 8.6, ROS 2 Humble; according to NVIDIA, all Orin modules are production-supported through 2032.
  • Carrier-board design: per NVIDIA design guidelines, AGX Orin requires 10+ PCB layers, PCIe Gen 4 length match under 5 mils. The LPDDR5 memory is soldered on the module, so the carrier board routes no DRAM bus: its critical links are PCIe, USB 3.x, MIPI CSI-2 and Ethernet.
  • Standards in scope: CE marking (EN 55032 Class A/B, EN 55035), FCC Part 15 Subpart B, IEC 60068-2-6 vibration, IPC-2221 for board design.
  • Field result: in Q1 2026 we delivered a Jetson Orin NX project with a fully custom Yocto Board Support Package (BSP).

What is the NVIDIA Jetson platform?

The NVIDIA Jetson platform is a family of System on Module (SoM) products designed for edge AI inference. Each module combines an NVIDIA Graphics Processing Unit (GPU) with Tensor Cores, a multi-core Arm Central Processing Unit (CPU) and high-bandwidth memory, so complex Neural Network (NN) models can run locally without a cloud link.

The software stack is one of the platform's strongest assets. According to NVIDIA, the JetPack 6.0 Software Development Kit (SDK) provides a complete development environment: Ubuntu 22.04 Long-Term Support (LTS), GPU drivers, CUDA 12.2, cuDNN, TensorRT 8.6 and optimised deployment tools. Per Canonical release notes, the Ubuntu 22.04 base gets security patches through April 2032, aligning well with industrial product lifecycles. This software maturity lets engineers move models from an NVIDIA workstation to an embedded module with little friction, cutting development time substantially. Frameworks such as PyTorch and TensorFlow export to ONNX Runtime or TensorRT for optimised inference.

Unlike cloud-dependent solutions that introduce network latency and raise data-privacy concerns, Jetson modules process information locally. This delivers real-time operation, full independence from connectivity, and end-to-end control over sensitive data, non-negotiable requirements for industrial, medical and defence applications.

Jetson lineup: module comparison

The Jetson lineup spans a wide performance range, from the entry-level Jetson Orin Nano up to the Jetson AGX Orin for the most demanding workloads. The module choice directly drives inference capability, power budget, and final system cost.

The table below compares four modules: the Jetson Orin Nano (40 TOPS, 7-15 W), the Jetson Orin NX (70-100 TOPS, 10-25 W), the Jetson AGX Orin (200-275 TOPS, 15-60 W) and the legacy Jetson Xavier NX (21 TOPS, 10-20 W). All three Orin modules use Arm Cortex-A78AE cores and LPDDR5 memory. For new projects, we recommend the Orin series over the Xavier family, which NVIDIA still supports for legacy designs, to benefit from the Ampere architecture and long-term software support through JetPack 6.x.

Eight NVIDIA Jetson Orin modules on the AESTECHNO qualification bench, heatsinks and fans visible, staged for thermal and IO characterisation.
NVIDIA Jetson Orin modules in qualification at the AESTECHNO Montpellier lab, prior to industrial carrier-board integration.
Module GPU CUDA Cores CPU RAM AI performance (TOPS) TDP Use cases
Jetson Orin Nano 1024 ARM Cortex-A78AE 4/8 GB LPDDR5 40 7-15 W Entry-level vision, educational robotics
Jetson Orin NX 1024 ARM Cortex-A78AE 8/16 GB LPDDR5 70-100 10-25 W Robotics, drones, AMR
Jetson AGX Orin 2048 ARM Cortex-A78AE 32/64 GB LPDDR5 200-275 15-60 W Autonomous vehicles, heavy industry
Jetson Xavier NX (legacy) 384 ARM Carmel 8/16 GB LPDDR4x 21 10-20 W Existing projects, migration

The Orin series is a generational leap over the Xavier family. The Jetson Orin Nano nearly doubles the AI performance of the Xavier NX in a more compact, more affordable footprint. NVIDIA still supports Xavier modules for legacy designs, but for new projects we recommend planning the migration to Orin to benefit from the Ampere architecture, LPDDR5 memory (102 GB/s vs 51 GB/s on Xavier LPDDR4x), and long-term software support through JetPack 6.x. As NVIDIA documentation notes, the Jetson Orin Nano Super refresh now reaches 67 TOPS in a 7-25 W envelope, a 1.7x boost over the original Orin Nano.

Jetson carrier board: the five critical disciplines Diagram of a carrier board around the Jetson Orin module with the five design disciplines: high-speed routing, stack-up, power and sequencing, thermal and EMC, with the figures quoted in the article. Jetson carrier board: five critical disciplines what the carrier board must guarantee around the module Carrier board Jetson Orin module (SoM) High-speed routing PCIe Gen 4 (16 GT/s), USB 3.x 85 / 90 / 100 ohm differential pairs intra-pair skew under 5 mils Stack-up AGX Orin: 10 layers minimum continuous reference plane under every differential pair Power and sequencing rails 5 / 3.3 / 1.8 / 0.8 V strict power-up order supervisor + controlled reset Thermal AGX Orin: up to 60 W sealed IP65/IP67: conduction to the chassis, simulated first EMC EN 55032 Class B, EN 55035 IEC 61000-4-2: 8 kV contact, 15 kV air filtering and shielding from layout Constraints quoted in the article: NVIDIA design guidelines, IEC 61000-4-2.
Figure: Jetson carrier-board design: the five disciplines our checklist covers around the module, from high-speed routing and stack-up to power sequencing, thermal management and EMC.
NVIDIA Jetson Orin lineup positioned on TOPS vs power consumption Two-axis chart positioning the three NVIDIA Jetson Orin modules: x-axis power consumption in watts (5 to 60 W), y-axis INT8 AI performance in TOPS (40 to 275). Orin Nano sits at 40 TOPS / 7-15 W for entry-level vision, Orin NX at 70-100 TOPS / 10-25 W for robotics and AMR, AGX Orin at 200-275 TOPS / 15-60 W for autonomous vehicles and heavy industry. Three coloured zones map to use-case clusters: edge vision, robotics, autonomous platforms. Jetson Orin lineup: TOPS versus power envelope INT8 inference performance against thermal budget: pick the module that matches the carrier cooling envelope Power envelope (W, TDP) 7 W 15 W 25 W 40 W 60 W AI performance (TOPS INT8) 0 40 100 200 275 Orin Nano 40 TOPS · 7-15 W edge vision, education Orin NX 70-100 TOPS · 10-25 W robotics, AMR, drones AGX Orin 200-275 TOPS · 15-60 W autonomous vehicles, heavy industry Heuristic ~ 5 TOPS / W on Orin family Source: NVIDIA Jetson Orin datasheets, JetPack 6.0 documentation. Production support through 2032 across the Orin family.
NVIDIA Jetson Orin family positioned on a TOPS-vs-power chart. The diagonal slope is roughly 5 TOPS per watt across the Orin family, so a power budget pins down the achievable inference performance before you ever pick the module.

Technical architecture

Jetson modules integrate several specialised accelerators inside a single System on Chip (SoC). The NVIDIA GPU with Tensor Cores, the 64-bit Arm CPU, the NVIDIA Deep Learning Accelerator (NVDLA) and the high-bandwidth memory together form an architecture optimised for high-performance, energy-efficient embedded inference. According to Arm, the Cortex-A78AE cores used in Orin are automotive-enhanced, with Dual-Core Lock-Step (DCLS) support for ISO 26262 ASIL-D compliance.

The list below details each building block: the GPU with Tensor Cores, the 64-bit Arm CPU, the NVDLA, the LPDDR4x or LPDDR5 memory, the PCIe Gen 4 bus, the MIPI CSI-2 camera interfaces and the general connectivity. For a carrier-board designer, the memory, PCIe and camera interfaces are the ones that set the routing constraints, with up to 102 GB/s of memory bandwidth and up to 16 GT/s per PCIe lane.

  • NVIDIA Volta/Ampere GPU with Tensor Cores: the core compute engine. Tensor Cores Gen 3 (on AGX Orin: 64 cores) accelerate the matrix operations of neural networks (convolutions, transformers), delivering inference performance well beyond a generic GPU. Per NVIDIA specs, sparse INT8 throughput on AGX Orin reaches 275 TOPS.
  • 64-bit Arm v8.2 CPU: Arm Cortex-A78AE (Orin) or Carmel (Xavier) cores handle the Operating System (OS), task orchestration and pre/post-processing. The big.LITTLE architecture optimises power as a function of workload.
  • NVIDIA Deep Learning Accelerator (NVDLA): a dedicated inference accelerator that runs in parallel with the GPU to raise overall throughput or lower power on standard inference tasks.
  • LPDDR4x / LPDDR5 memory: memory bandwidth is critical to AI performance. Orin modules use LPDDR5 with up to 102 GB/s versus 51 GB/s for Xavier's LPDDR4x. That memory is soldered on the module itself: NVIDIA has already routed it, and the carrier-board designer never touches the DRAM bus.
  • PCIe Gen 4 bus: this high-speed link connects expansion peripherals (NVMe storage, network cards, additional accelerators) at up to 16 GT/s per lane.
  • Camera interfaces (MIPI CSI-2): simultaneous multi-stream video, essential for multi-camera vision applications (up to 6 cameras on AGX Orin).
  • Connectivity: USB 3.2, Gigabit Ethernet, GPIO, I2C, SPI, UART, a complete set for integration into complex industrial systems.

Designing a Jetson carrier board

The carrier board is the host PCB that mounts the Jetson module and provides connectors, power and application-specific interfaces. Its design is a major engineering exercise, combining high-speed routing, thermal management and EMC compliance, and directly shapes the system's performance and reliability.

Field note, Jetson Orin NX project (Q1 2026). In Q1 2026 we delivered a complete project on a Jetson Orin NX module with a fully custom Board Support Package (BSP) developed under Yocto: custom layers, kernel configuration, device tree and rootfs. A demanding project, from which we drew significant lessons.

In our practice at AESTECHNO we have found that a successful Jetson carrier board design depends on simultaneous mastery of several critical disciplines. Unlike generic SBC designs, Orin carrier boards place signal integrity on the critical path from day one. Our approach differs from reference designs in that we simulate every critical bus in ANSYS SIwave before the first PCB spin.

At AESTECHNO the carrier board checklist covers these disciplines:

  • High-speed routing (PCIe, USB 3.x, MIPI CSI-2): the carrier board's PCIe Gen 4 links, USB 3.x, MIPI CSI-2 and Ethernet pairs demand strict control of differential impedance (85 Ω for PCIe, 90 Ω for USB, 100 Ω for MIPI CSI-2 and Ethernet), intra-pair skew (under 5 mils) and trace length. A few millimetres off can erode timing margin and cause intermittent errors that are very hard to diagnose. Per IPC-2141A guidelines, stripline impedance tolerance should stay within ±10%.
  • Stack-up and impedance control: a tailored PCB stack-up is essential. For an AGX Orin carrier board we recommend a minimum of 10 layers with continuous reference planes under every differential pair.
  • Multi-rail power supply and sequencing: Jetson modules require several supply voltages (5 V, 3.3 V, 1.8 V, 0.8 V) brought up in a precise order. Incorrect sequencing can damage the module or cause erratic boots. The power management design must use low-noise regulators and careful decoupling.
  • Thermal management: a Jetson AGX Orin can dissipate up to 60 W under maximum AI load. In a sealed enclosure (IP65/IP67), natural convection alone is insufficient. We design conduction-based cooling solutions to a metal chassis, with prior thermal simulation to guarantee operation across the industrial temperature range.
  • Electromagnetic compatibility (EMC): the high frequencies of high-speed buses generate significant emissions. EMC compliance for CE marking (EN 55032 Class B, EN 55035, EN 61000-4-2/4-3/4-4/4-5/4-6) and FCC Part 15 Subpart B certification requires rigorous shielding, filtering and routing of critical traces from the design phase. According to IEC 61000-4-2, the Electrostatic Discharge (ESD) test applies ±8 kV contact and ±15 kV air discharge.
JetPack software stack and AI model path to inference Five-layer stack: Tegra Orin hardware with Ampere GPU and NVDLA, Linux 6.1 LTS kernel and Tegra drivers, JetPack 6 SDK with CUDA 12.2 cuDNN TensorRT 8.6 DeepStream, frameworks PyTorch TensorFlow ROS 2, embedded application. On the right, an INT8-quantised YOLOv8 inference example. NVIDIA JetPack 6 software stack on Jetson Orin From Tegra silicon to AI application, NVIDIA production support through 2032 Layer 1 Hardware, Tegra Orin SoC Cortex-A78AE CPU · Ampere GPU + Tensor Cores Gen3 · NVDLA · LPDDR5 Layer 2 Linux 6.1 LTS + Tegra drivers (L4T) Ubuntu 22.04 LTS or custom Yocto · DTB · CBoot bootloader · UEFI Layer 3 JetPack 6 SDK CUDA 12.2 · cuDNN · TensorRT 8.6 · DeepStream · VPI · Multimedia API Layer 4 AI frameworks + middleware PyTorch · TensorFlow · ONNX Runtime · ROS 2 Humble · Triton Inference Layer 5 Embedded application Industrial vision, SLAM, sensor fusion, robot control, NGC containers AI model path YOLOv8n, object detection PyTorch FP32 (.pt) trained on NVIDIA workstation ONNX export portable graph TensorRT 8.6 (.engine) INT8 quantisation + fusion Tensor Cores inference throughput gain vs FP32 DeepStream pipeline CSI-2 camera → ISP → CUDA → TRT Product life cycle aligned with Ubuntu 22.04 LTS (security patches through April 2032, per Canonical release notes) and the JetPack 6.x roadmap published by NVIDIA. Source: NVIDIA Jetson Linux Developer Guide, JetPack 6.x release notes.
Figure: JetPack 6 software stack on Tegra Orin: five layers from silicon to application, on the right the path of a YOLOv8 model from PyTorch to INT8 inference on Tensor Cores.

Industrial applications of embedded AI

Embedded AI on the Jetson platform fits any sector that needs intelligent real-time data processing. From industrial vision to autonomous robotics, these modules let teams deploy complex inference models directly in the field, no cloud infrastructure required.

Six application families stand out: industrial vision and quality control, autonomous robotics and AMR, autonomous vehicles, intelligent video surveillance, medical and diagnostic imaging, and precision agriculture. The module choice follows the workload: the Jetson Orin NX offers the best performance/power trade-off for mobile robots, while the Jetson AGX Orin, with 275 TOPS, handles multi-sensor fusion for autonomous vehicles. In every case, local processing removes the cloud dependency and keeps sensitive data on the device, from the factory floor to the hospital.

  • Industrial vision and quality control: defect detection on high-speed production lines. Classification and segmentation models run in real time (30–60 FPS) to flag non-conforming parts without slowing the line.
  • Autonomous robotics and AMR: Autonomous Mobile Robots use Jetson modules for SLAM navigation, obstacle avoidance and path planning. The Jetson Orin NX offers the best performance/power trade-off for this category.
  • Autonomous vehicles: the Jetson AGX Orin, with 275 TOPS, enables multi-sensor fusion (cameras, LiDAR, radar) and real-time perception. It is the reference platform for prototyping and pre-production of autonomous vehicles.
  • Intelligent video surveillance: multi-stream video analytics with event detection, behaviour recognition and people counting. NVIDIA's DeepStream framework optimises the end-to-end video pipeline.
  • Medical and diagnostic imaging: processing of medical images (radiology, dermatology, ophthalmology) with diagnostic-aid AI models. Local processing keeps patient data confidential, in line with healthcare regulations.
  • Precision agriculture: drones equipped with multispectral cameras and Jetson modules for real-time crop health analysis, plant disease detection, and spraying optimisation.

Jetson vs alternatives: how to choose

Picking an embedded AI platform depends on multiple criteria: required inference performance, software ecosystem, time-to-market, solution flexibility, and target production volumes. Below we compare the four main approaches to help decision-makers identify the right fit.

The four approaches are the NVIDIA Jetson, a custom FPGA, a dedicated ASIC and the Raspberry Pi. The Jetson combines very high AI performance with a mature CUDA/TensorRT ecosystem and a time-to-market of months, for volumes of 100 to 10k units per year. A custom FPGA offers very high flexibility in 6 to 12 months, a dedicated ASIC delivers maximum optimisation above 100k units per year but takes years to develop, and the Raspberry Pi suits prototyping with CPU-only AI performance.

Criterion Jetson (NVIDIA) Custom FPGA Dedicated ASIC Raspberry Pi
AI performance Very high (GPU + Tensor Cores) High (programmable) Maximum (optimised) Low (CPU only)
Software ecosystem CUDA / TensorRT mature Specialised HDL Proprietary SDK Generic Linux / Python
Time-to-market Fast (months) Medium (6-12 months) Long (years) Very fast
Flexibility High (reprogrammable) Very high (reconfigurable) Low (frozen) High
Power Moderate (7-60 W) Variable Optimal Low (5-15 W)
Optimal volume 100-10k units/year 1k-100k >100k Prototyping

For most industrial projects with volumes from a few hundred to a few thousand units per year, the Jetson remains the most rational choice. The CUDA/TensorRT ecosystem moves a prototype to production in months, where a custom FPGA or ASIC requires significantly higher development investment.

Common pitfalls and field lessons

Integrating a Jetson module into an industrial product brings specific technical challenges that only field experience surfaces. At AESTECHNO we have supported engineering teams on embedded AI projects, including a Jetson Orin NX project, and we share below the most frequent pitfalls and the proven solutions. They are thermal under-sizing, carrier-board routing errors, JetPack version incompatibilities, power-sequencing errors and EMC non-compliance during certification.

Recent client project. In Q1 2026 we delivered a Jetson Orin NX project with a fully custom Yocto BSP. In parallel, we supported a client through the industrialisation of a product built around a high-power AI ASIC, an alternative to Jetson modules when volumes and power constraints justify dedicated silicon. Both engagements feed directly into the recommendations below.

The five most frequent pitfalls

Thermal under-sizing. At AESTECHNO this is the most frequent error we see. A Jetson AGX Orin pulling 60 W at maximum AI load inside a sealed IP67 enclosure generates considerable thermal density. Without prior thermal simulation and a suitable dissipation strategy (chassis conduction, heat pipes, forced air), the module triggers thermal throttling and performance collapses. We systematically run a Computational Fluid Dynamics (CFD) thermal simulation before fabricating the first prototype, validating the design against IEC 60068-2-2 dry-heat and IEC 60068-2-14 thermal-cycling profiles.

Carrier-board PCB routing errors. DDR and PCIe buses demand rigorous control of differential impedance and length matching. We have seen projects where a few-millimetre mismatch on PCIe Gen 4 differential pairs caused intermittent link errors that were extremely difficult to diagnose in production. Strict adherence to NVIDIA's design guidelines and the use of signal-integrity simulators are non-negotiable.

JetPack version incompatibilities. The JetPack SDK evolves regularly, and updates can introduce incompatibilities with camera drivers, TensorRT libraries or middleware layers. According to the NVIDIA Developer Forums, most regression tickets on JetPack 6.0 involve MIPI CSI-2 drivers or Deep Learning Accelerator (DLA) quantisation. Lock the JetPack version at the start of the project and plan version migrations through a dedicated test and validation cycle.

Power-sequencing errors. Jetson modules require a strict power-up order across rails. Incorrect sequencing can cause destructive latch-up or erratic boots. The power-management circuit must follow NVIDIA's recommendations and include voltage supervision with a controlled reset.

EMC non-compliance during certification. The high frequencies of high-speed links (PCIe Gen 4 at 8 GHz, USB 3.x, MIPI CSI-2) generate emissions that can exceed CE/FCC limits if shielding and filtering are not properly sized. We integrate EMC constraints from the placement-and-routing phase to avoid expensive respins after lab testing.

NVIDIA Jetson: a strategic lever for embedded AI

Embedded AI is no longer a technology promise, it is a measurable competitive advantage. At AESTECHNO we work with technical directors and innovation leads on edge-AI integration, and we have observed that the choice of hardware platform directly shapes project success.

Three angles structure this platform choice for a technical director: the competitive advantage of edge AI, which brings reduced latency, data privacy and autonomous operation; the comparison between Jetson and a custom ASIC, where Jetson wins on time-to-market for a few hundred to a few thousand units per year; and the positioning in Industry 4.0, where intelligence is deployed as close as possible to machines for vision-based quality control, predictive maintenance and real-time process optimisation.

The competitive advantage of edge AI

Processing data locally, without cloud dependency, yields three strategic benefits: reduced latency (real-time guaranteed), data privacy (no network transit) and autonomous operation (no permanent connection required). For industrial, medical or defence sectors, these are not nice-to-haves, they are prerequisites. Thanks to their optimised GPU architecture and high-bandwidth LPDDR memory, Jetson modules run complex inference models directly in the field.

Jetson vs custom ASIC: time-to-market and risk

The main alternative to Jetson for embedded AI remains a custom ASIC or FPGA design. Jetson has a decisive time-to-market advantage: NVIDIA's software ecosystem (CUDA, TensorRT, JetPack) takes a project from prototype to production in months, where an ASIC requires years of development. An ASIC, in return, offers maximum optimisation at high volume. For most industrial projects (a few hundred to a few thousand units per year), Jetson is the most rational choice.

Industry 4.0 positioning

Industry 4.0 hinges on deploying intelligence as close as possible to machines: vision-based quality control, predictive maintenance, real-time process optimisation. Jetson modules slot naturally into this architecture thanks to their connectivity (PCIe, USB, Ethernet, CSI camera interfaces) and their compatibility with industrial standards. The choice of memory technology (DDR4/DDR5) and the security of AI systems are aspects we integrate from the architecture phase to ensure long-term solution viability.

Embedded AI project with Jetson? AESTECHNO expertise

Building a vision/AI system on NVIDIA Jetson? Our engineers help you with:

  • Custom carrier board design (Orin Nano, Orin NX, AGX Orin)
  • Multi-channel camera and sensor integration
  • CUDA/TensorRT application development
  • Thermal management and industrialisation

30-minute audit (free)

Free resource: our electronics specification template

Why choose AESTECHNO?

  • 10+ years of experience in electronic design
  • 100% success rate on CE/FCC certifications, across dozens of campaigns run over more than 10 years of practice
  • French design firm based in Montpellier

Article written by Hugues Orgitello, electronic design engineer and founder of AESTECHNO. LinkedIn profile.

Related

FAQ: Jetson processors

What is the difference between Jetson Nano, Xavier NX, Orin and AGX Orin?
Jetson Nano: entry-level, 128 CUDA cores, 4 GB RAM, 472 GFLOPS FP16, ~10 W, ideal for prototyping and education. Xavier NX: mid-range, 384 CUDA cores, 8-16 GB, 21 TOPS, 15 W, robotics and drones. Orin Nano: 1024 CUDA cores, 8 GB, 40 TOPS, 15 W, new generation. AGX Orin: high-end, up to 2048 CUDA cores, 64 GB, 275 TOPS, 60 W, autonomous vehicles and industry. Choose based on AI inference needs and budget.

Jetson vs Raspberry Pi: when do you pick Jetson?
Raspberry Pi: generic CPU (ARM Cortex), no dedicated AI GPU, low cost but unsuited to heavy inference (object detection above 5 FPS is hard). Jetson: NVIDIA GPU with AI-optimised Tensor cores, CUDA/cuDNN/TensorRT acceleration, real-time inference (30-60 FPS on complex neural networks). Pick Jetson for: real-time computer vision, embedded deep learning, autonomous robotics. Keep Raspberry Pi for: generic applications without heavy AI workloads.

Which AI frameworks are supported on Jetson?
Native support: TensorFlow, PyTorch, ONNX, TensorRT (NVIDIA optimisation), DeepStream (video pipelines), CUDA, cuDNN. The Jetson Inference library provides pre-trained models (classification, detection, segmentation) ready to use. The JetPack SDK bundles Ubuntu, GPU drivers, AI libraries and development tools. Excellent compatibility with the standard Python/AI ecosystem and easy migration from an NVIDIA workstation.

How do you handle thermal dissipation on Jetson modules?
Jetson Nano/Orin Nano (10-15 W): a passive heatsink suffices in a ventilated enclosure. Xavier NX (15-20 W): a fan or oversized heatsink is recommended. AGX Orin (30-60 W): active cooling is mandatory (fan + copper heatsink). For harsh industrial environments: IP65 enclosures with conduction cooling to an external metal chassis. Automatic thermal throttling protects the SoC but degrades performance, size the thermal solution properly.

Can Jetson be used in harsh industrial environments (-40 °C to +85 °C)?
Yes, with a Jetson AGX Orin Industrial module (extended range -40 °C to +85 °C) and a tailored design. Precautions: active cooling/heating depending on ambient temperature, stabilised power supply (industrial over/under-voltage protection), industrial eMMC flash storage (extended write cycles), conformal coating (humidity/dust protection).