This page was automatically translated and may contain errors. View in English.
O
Stage

Computer Vision Embedded AI Internship

OMNIDYA TECH LLP

Ahmedabad, Gujarat, India Stage à temps plein

Soyez le premier à postuler

Traitement
Allocation : 12 000 à 15 000 INR par mois
Durée
6 mois
Commencer
Immédiatement
Ouvertures
2
Qui peut postuler

Candidates eligible to apply must be able to undertake a full-time, on-site internship located in Ahmedabad, with availability to start between July 13 and August 17, 2026. The candidate must commit for a six-month period and have the pertinent skills and academic background relevant to computer vi…

Mode de travail
Au bureau
Éducation
B.E./B.Tech computer science or related
CV
Candidature requise

À propos du stage

About the Internship

This internship offers hands-on experience in training, fine-tuning, and assessing object detection and image classification models tailored for both road-facing and cabin-facing camera systems. You'll contribute to building and managing image datasets by aiding in data annotation, class balancing, and mine hard-negative samples from real-world footage.

Evaluating model effectiveness using metrics like mean Average Precision (mAP), Intersection over Union (IoU), precision, and recall is a key responsibility, including analyzing failures across varied lighting, weather, and occlusion scenarios. You'll also assist with optimizing models, conducting quantization, and deploying them on edge AI hardware while measuring accuracy and inference speed.

Working intensively with OpenCV, Python, and deep learning frameworks will form a core part of your role as you develop and validate innovative computer vision solutions. Supporting experiments involving camera tuning, image processing, and sensor fusion leveraging data from camera, LiDAR, and IMU sensors is expected. Additionally, collaboration with Embedded AI and Firmware teams to document work, enhance performance, and ensure reproducibility will be essential.

Candidate Eligibility

  • Applicants must be available for a full-time, in-office internship located in Ahmedabad.
  • Internship start date should be between July 13, 2026, and August 17, 2026.
  • The internship duration is six months with full commitment required.
  • Applicants should possess relevant skills and interests aligned with the role.
  • Open for Computer Science Engineering students and fresh graduates seeking employment.

Required Qualifications and Skills

  • Currently pursuing or recently completed B.E./B.Tech/M.Tech/M.Sc. in Computer Science, Electronics & Communication, Electrical Engineering, Artificial Intelligence, Data Science, or a related discipline.
  • Practical experience with Python, OpenCV, NumPy, and either PyTorch or TensorFlow from academic projects, internships, research initiatives, or competitions.
  • Familiarity with Linux, Git, fundamental computer vision and object detection principles.
  • Additional exposure to YOLO, TensorFlow Lite, ONNX, TensorRT, CVAT, Roboflow, or automotive vision datasets is advantageous.

Compensation and Benefits

  • Monthly stipend ranging from INR 12,000 to 15,000.
  • A certificate upon completion to validate your experience.
  • Letter of recommendation acknowledging your contributions.
  • Opportunity for a full-time job offer after successful internship completion.
  • A comfortable work environment with informal dress code and a five-day work week.

Additional Details

This internship is hosted by an Ahmedabad-based AI technology company specializing in dashcam technologies. The firm operates as the Indian branch of a well-established US-based entity.

On transitioning to permanent employment, candidates can expect an annual salary between INR 400000 and 500000.

Avantages

Certificat Lettre de recommandation Offre de pré-embauche (PPO) semaine de 5 jours

Laissez ce message si vous souhaitez une réponse — nous ne l'utiliserons à aucune autre fin.

Cliquez pour parcourir, glisser-déposer, ou coller une capture d'écran

PNG, JPG, GIF, MP4, WebM, MOV · 20 Mo maximum par fichier · Jusqu'à 5 fichiers

🤖
En ligne · Aide IA instantanée