ADAS & Robot

ADAS & Robot ADAS & Robot Viet

[VIETTEL] CHIÊU MỘ KỸ SƯ AI, NPU, RTL, LẬP TRÌNH NHÚNG
11/04/2026

[VIETTEL] CHIÊU MỘ KỸ SƯ AI, NPU, RTL, LẬP TRÌNH NHÚNG

👋 (F-Automotive) Calling Technical Lead (MCAL/AUTOSAR) – Signing Bonus + Relocation Support 🚗Bạn đang làm Automotive và ...
11/04/2026

👋 (F-Automotive) Calling Technical Lead (MCAL/AUTOSAR) – Signing Bonus + Relocation Support 🚗
Bạn đang làm Automotive và mong muốn:
👉 Làm việc trực tiếp với khách hàng EU?
👉 Lead team, tăng ownership thay vì chỉ code?
👉 Tham gia dự án chuẩn quốc tế (AUTOSAR, ECU…)?
Đây là cơ hội dành cho bạn.
💼 Role highlights:
Kỹ năng giao tiếp tiếng Anh tốt (nói và viết).
Kinh nghiệm thực tế với MCAL/Autosar.
Kinh nghiệm code review, estimation & tham gia trao đổi kỹ thuật trực tiếp với khách hàng.
Có kinh nghiệm dẫn dắt team
💰 Benefits:
🎁 Signing bonus available (theo level)
Or 🌊 Relocation support 80–100M về Đà Nẵng
💸 Competitive salary + project bonuses
📍 Location: FPT Complex, Đà Nẵng (onsite)

[F-Automotive] Chill chill cũng 2 vị trí technical Lead SA Embedded/MCAL/ Autosar.Xua tan cái nóng cũng hỗ trợ chuyển vù...
11/04/2026

[F-Automotive] Chill chill cũng 2 vị trí technical Lead SA Embedded/MCAL/ Autosar.
Xua tan cái nóng cũng hỗ trợ chuyển vùng về biển xanh ĐN, trọn gói 100M
---ib mình hỗ trợ liền tay nhé--
👋 (F-Automotive) Calling Technical Lead (MCAL/AUTOSAR) – Signing Bonus + Relocation Support 🚗
Bạn đang làm Automotive và mong muốn:
👉 Làm việc trực tiếp với khách hàng EU?
👉 Lead team, tăng ownership thay vì chỉ code?
👉 Tham gia dự án chuẩn quốc tế (AUTOSAR, ECU…)?
Đây là cơ hội dành cho bạn.
💼 Role highlights:
Kỹ năng giao tiếp tiếng Anh tốt (nói và viết).
Kinh nghiệm thực tế với MCAL/Autosar.
Kinh nghiệm code review, estimation & tham gia trao đổi kỹ thuật trực tiếp với khách hàng.
Có kinh nghiệm dẫn dắt team
💰 Benefits:
🎁 Signing bonus available (theo level)
Or 🌊 Relocation support 80–100M về Đà Nẵng
💸 Competitive salary + project bonuses
📍 Location: FPT Complex, Đà Nẵng (onsite)

🚨 [URGENT] EMBEDDED SOFTWARE ENGINEER (AUTOSAR,C++)📍 Địa điểm: Hồ Chí Minh City✅ YÊU CẦU:2+ năm trong mảng EmbeddedThành...
11/04/2026

🚨 [URGENT] EMBEDDED SOFTWARE ENGINEER (AUTOSAR,C++)
📍 Địa điểm: Hồ Chí Minh City
✅ YÊU CẦU:
2+ năm trong mảng Embedded
Thành thạo C++, AUTOSAR
Có kinh nghiệm với MCU
✅ QUYỀN LỢI:
💰 13th Salary + Performance Bonus
🏥 Premium Healthcare (PVI) & family benefits
⏰ Flexible working time – 8h onsite linh hoạt
🌴 17 ngày nghỉ/năm (12 phép + 5 sick leave)
Nhận JD & ứng tuyển ngay:
📧 Email: [email protected]
📱 Zalo: 0377 001 806
👉 Tag bạn bè phù hợp hoặc inbox để được tư vấn chi tiết!

11/04/2026

VinFast - Viện Phần mềm dịch vụ thông minh tuyển dụng
1. 𝐀𝐧𝐝𝐫𝐨𝐢𝐝 𝐃𝐞𝐯𝐞𝐥𝐨𝐩𝐞𝐫: từ 3 năm kinh nghiệm, 1 vòng pv online
2. 𝐒𝐞𝐧𝐢𝐨𝐫 𝐁𝐚𝐜𝐤𝐞𝐧𝐝 𝐄𝐧𝐠𝐢𝐧𝐞𝐞𝐫 (𝐉𝐚𝐯𝐚 𝐒𝐩𝐫𝐢𝐧𝐠 𝐁𝐨𝐨𝐭): từ 5 năm kinh nghiệm, 1 vòng pv online
3. 𝐃𝐚𝐭𝐚 𝐄𝐧𝐠𝐢𝐧𝐞𝐞𝐫: từ 3 năm kinh nghiệm, 1 bài test + 1 vòng pv online
Địa điểm làm việc: Technopark, Oceanpark, Gia Lâm
JD và link ứng tuyển dưới comment! 👇
Các bạn quan tâm job vui lòng ứng tuyển qua link hoặc gửi CV về địa chỉ mail [email protected] nhé!

🤑 The software platform and vehicle OS race is on! As always, please add your thoughts 💭 below. Just sharing my personal...
14/12/2025

🤑 The software platform and vehicle OS race is on!

As always, please add your thoughts 💭 below. Just sharing my personal opinion.

👉 OEMs are in early stages of shifting their portfolio to SDVs, while selective Chinese and American OEMs are already on the AI-defined vehicle S-Curve.

👉 Abstraction of HW & SW is seen as the key enabler for max flexibility on the HW side and as leverage customization and localization on the application layer.

👉 This shift requires new software platforms / vehicle OSes to fill the layer between (compute) hardware and application software.

👉 Suppliers are fighting for the foundational layer, OEMs are turning into platform suppliers and OSS initiatives want to win with vendor-agnostic vehicle OS solutions.

👉 Reduced R&D spendings 🤑, lower maintenance costs and faster time to market are the buzz words that all camps are claiming.

Let’s dive into the different ecosystems:

1️⃣ Supplier camp is getting ready:

🔘 Suppliers and 3rd party vendors like QNX, TTTech Auto & VECTOR Informatik join forces to create a foundational SW platform for the SDV players.

🔘 Foxconn is working on multiple vehicle OSes, e.g., together with Elektrobit. Reference platforms could be used by existing OEMs and new players.

🔘 Huawei is the ultimate player in China 🇨🇳, multiple OEMs and new brands utilize the SW platforms to scale their SDV portfolio. It’s a benchmark for scale and time to market 🚀.

2️⃣ OEM camp: Becoming platform suppliers

🔘 New OEMs like XPENG, Rivian or NIO have invested huge R&D amounts to create the modern E/E architectures and SW platforms to power SDVs and AIDVs.

🔘 Volkswagen Group, Stellantis or smaller OEMs like Forseven are leveraging the smart vehicle platforms from XPENG, Rivian, Nio or Leapmotor.

🔘 Volkswagen and Rivian formed the Rivian and Volkswagen Group Technologies JV to develop solutions for Rivian, Volkswagen and 3rd party OEMs according to key executives: https://lnkd.in/dw9jKeEu

🔘 Renault Group and Volvo Cars are joining the camp: Ford Motor Company is using the Ampere platform for next gen EVs in Europe. Volvo is open for business to license the latest tech stack. Give Anders Bell a call ☎️ if you are interested.

3️⃣ Vendor-agnostic Open-Source ecosystem

🔘 Li Auto joined the Open-Source camp with Halo OS. It's the OSS version of the own SW platform, covering all domains on application layers, except IVI OS.

🔘 Multiple OEMs & suppliers joined forces to develop S-CORE, a vendor agnostic and non-differentiating middleware.

⚠️ It’s still a long way to profitable and sustainable businesses:

🔘 The margins are thin, SaaS and license prices are low 😬.

🔘 Previous attempts from ETAS or Microsoft and others failed.

🔘 To be announced platforms by Tier-1s or service providers heat up the competition.

🔘 OEMs might prefer the vertical integration, due to internal political reasons, control points & velocity.

🔘 It still open if Open-Source is a quiet kingmaker.

🚗Restricted Autonomy: The First Milestone Toward Self-Driving CarsWhen we think of self-driving cars, we imagine vehicle...
14/12/2025

🚗Restricted Autonomy: The First Milestone Toward Self-Driving Cars

When we think of self-driving cars, we imagine vehicles navigating effortlessly on their own.
But the journey actually began with a very early stage called Restricted Autonomy - a system that could assist the driver, but only within strict limits.

How It Worked: 3D Maps + Sensors
Restricted autonomy introduced something new for the automotive world:
cars that could sense, understand, and react.

🔹 High-Definition 3D Maps
Cars used detailed maps showing lane boundaries, curves, slopes, speed limits, and road structures. This told the vehicle where it was and what to expect next. But these maps worked only in pre-mapped areas making the system “restricted.”

🔹 Sensors Around the Vehicle
Cameras for lanes and signs
Radar for distance and speed
Ultrasonic sensors for close obstacles
Sometimes LiDAR for depth
These gave the car a 360° view of its environment.

🔹 Together
Maps provided the planned layout.
Sensors provided real-time reality.

🚫 But Why Didn’t It Work Long-Term?
Restricted autonomy was a breakthrough, but it couldn’t evolve into true self-driving. Here’s why:
1. It depended too much on pre-mapped areas
2. It struggled with real-world unpredictability
3. It didn’t learn in real time

This early stage didn’t create self-driving cars.
But it sparked the revolution that eventually will.

Nissan Motor Co. is making its boldest move yet to challenge Tesla in autonomous driving.The Japanese automaker is devel...
14/12/2025

Nissan Motor Co. is making its boldest move yet to challenge Tesla in autonomous driving.

The Japanese automaker is developing a hands-off, eyes-on system modeled after Tesla’s supervised Full Self-Driving, with no geographic restrictions.

“There’s a renewed interest in the company to have a truly capable self-driving system,” Nissan Americas product planning chief Ponz Pandikuthira told Automotive News.

Nissan expects the first vehicles with its artificial intelligence-powered, next-generation ProPilot system to initially arrive in Japan and North America. Like the Tesla system, Nissan intends to use a camera-based approach that saves the cost of using lidar.

14/12/2025

🚀 Real-Time Traffic Sign Recognition for Driver Assistance (ADAS Prototype)

Proud to share a production-oriented prototype from my AI Projects series: a Real-Time Traffic Sign Recognition system built for low-latency driver-assistance workflows.

🛑 The Problem
Modern ADAS platforms need fast, reliable sign detection.
Key challenges include:
🔹 High-speed recognition of incoming signs
🔹 Low-latency inference on non-GPU hardware
🔹 Real-time AR feedback for enhanced driver awareness

💡 The Solution
Built a lightweight, real-time system using:
🟠 PyTorch CNN trained on GTSRB
🟠 OpenCV for live video capture + preprocessing
🟠 AR Overlay Engine to display class + confidence instantly
Final result: ~20–25 FPS on CPU with seamless augmented feedback.

⚙️ Pipeline
Camera Feed → Preprocess (32×32) → CNN Inference → Sign Classification → AR Overlay

🔥 What Makes It Unique
🌈 Real-time throughput on a standard CPU
🌈 Instant visual confirmation (class + confidence)
🌈 Optimized model footprint for deployment flexibility
🌈 Low-latency AR rendering for improved driver context

🛠️ Technical Challenges Solved
🔧 Kernel Crashes → Fixed via isolated Conda env (tsr_env) + clean CPU-only installs
🔧 We**am Init Failure → Resolved using cv2.CAP_DSHOW + index fallback (0/1/2)
🔧 Incorrect Label Mapping → Identified as UI label mismatch, model classification was correct

🧠 Key Learnings
🧩 Deployment matters as much as model accuracy
⚡ Optimized CNNs can deliver strong real-time performance on CPU
🔍 System integration (PyTorch + OpenCV) requires precise configuration

DAS teams know the struggle:High-quality LiDAR datasets are hard to annotate, expensive to scale, and absolutely mission...
14/12/2025

DAS teams know the struggle:
High-quality LiDAR datasets are hard to annotate, expensive to scale, and absolutely mission-critical for perception model performance.
That’s exactly the problem we’re solving at JTheta.ai.

In our latest blog, we break down:
• The technical challenges of LiDAR annotation
• How annotation quality directly impacts ADAS decision-making
• The role of 3D segmentation, bounding boxes, and multi-sensor fusion
• Why enterprise-scale workflows are essential for reliable autonomous systems

If you work in automotive AI, GeoAI, ADAS, or autonomous robotics, this article delivers insights you shouldn’t miss.
Read the full blog:
https://lnkd.in/eDap9HVD

hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag hashtag

Address

High Tech District 9
Ho Chi Minh City

Alerts

Be the first to know and let us send you an email when ADAS & Robot posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.

Share