GenAI with Automotive Experts

KPIT; Maharashtra; On-site; INR 1,800,000 - 2,800,000; Full time

Job summary

KPIT in Maharashtra, India is seeking a Multimodal GenAI Architect / AI-ML Engineer to design and implement cutting-edge multimodal models. You will lead data pipelines, model training, and deployment for production-ready AI systems.

Join a team focused on scalable architectures, continual optimization, and integration of vision-language and audio-text fusion models, with emphasis on latency, accuracy, and robust evaluation in real-world applications.

Qualifications

• Strong foundation in deep learning, LLM integration, and AI/ML frameworks (PyTorch).
• Hands-on experience fine-tuning foundation models.
• Experience with multimodal model architectures (vision-language, audio-text fusion).
• Understanding of evaluation, testing, and production deployment practices.

Responsibilities

• Develop multimodal architectures: design, train, and test AI models across modalities.
• Design data pipelines for collection, alignment, and normalization across modalities.
• Optimize models for performance, latency, and contextual accuracy in production.
• Adapt and fine-tune GenAI/foundation models to meet requirements.

Skills

GenAI

Deep Learning

LLM Integration

PyTorch

Fine-tuning Foundation Models

Multimodal Architectures

Model Evaluation

Production Deployment

Tools

MLOps Tools

Job description

Job Description

Automotive background experience, Multimodal GenAI Architect / AI-ML Engineer

Key Responsibilities

• Develop Multimodal Architectures: Design, train, and test AI models capable of processing, understanding, and fusing inputs from multiple modalities (text, image, audio, video, etc.).
• Data Pipeline Design: Build scalable pipelines for data collection, temporal alignment, and normalization across modalities.
• Model Optimization: Improve performance, latency, and contextual accuracy of multimodal models in production environments.
• Architecture Modification: Adapt and fine-tune existing GenAI/foundation model architectures to meet specific application requirements.

Required Skills

• Strong foundation in deep learning, LLM integration, and AI/ML programming frameworks (PyTorch).
• Hands‑on experience fine‑tuning foundation models.
• Experience with multimodal model architectures (e.g., vision‑language models, audio‑text fusion).
• Solid understanding of model evaluation, testing, and production deployment practices.

Nice to Have

• Experience with distributed training / large-scale model training.
• Familiarity with MLOPs tools for pipeline automation and model monitoring.

Required Skills: genai, ai/ml, Deep learning, LLM Integration, Pytrorch, fine-tuning foundation models. Supported Skills: Develop Multimodal Architectures, Data Pipeline Design, Model Optimization, Familiarity with MLOps tools.