Description

You will build production-grade ML and multimodal systems supporting media workflows like metadata extraction, video understanding, speech transcription, moderation, and recommendations.

Responsibilities

• Build and deploy multimodal ML models covering NLP, CV, ASR, OCR, and video analytics.
• Develop pipelines for video frame extraction, shot detection, speech-to-text, and content moderation.
• Implement Retrieval Augmented Generation (RAG) using multimodal indexing.
• Optimize model latency through batching, quantization, or GPU optimizations for real-time workflows.
• Build and maintain ML CI/CD pipelines and deploy models as microservices using Docker and Kubernetes.

Required Skills

• 5+ years of experience in ML engineering.
• Strong proficiency in Python and PyTorch/TensorFlow.
• Hands-on experience with NLP and Computer Vision models.
• Knowledge of multimodal LLM frameworks including LangChain, LLaVA, or LlamaIndex.
• Experience with vector search implementations (e.g., Pinecone, FAISS).
• Proficiency with containerization using Docker and orchestration with Kubernetes.
• Experience building CI/CD pipelines.

Key Skills

Education

Any Graduate