Job Description

Key Responsibilities

Develop and train deep learning models for object detection, tracking, and recognition in images and videos (e.g., YOLO, Faster R-CNN, Detectron2). Build pipelines for video frame processing, feature extraction, and indexing. Implement semantic and hybrid search systems using LLMs, embeddings, and vector databases. Integrate retrieval-augmented generation (RAG) workflows for context-aware, natural language search. Develop APIs and web services to connect detection outputs with the search system. Work with large-scale datasets and ensure scalability and efficiency of search pipelines. Collaborate with cross-functional teams (data engineers, frontend developers, product teams) to deliver production-ready solutions.

Required Skills:

3-5 Years of Professional Experience Strong coding experience in Python. Expertise in computer vision frameworks (OpenCV, PyTorch, TensorFlow, Detectron2, YOLO). Solid knowledge of object detection, segmentation, and video analytics. Hands-on experience with LLMs and NLP frameworks (LangChain, LlamaIndex, Hugging Face). Familiarity with vector databases and embeddings (FAISS, Pinecone, Weaviate, Milvus). Experience with backend Frameworks(NodeJS) Experience with Docker, Git, and CI/CD workflows.

Nice to Have:

Knowledge of geospatial/GIS systems (ArcGIS, QGIS, PostGIS, Mapbox). Experience with multi-modal search (text + image/video). Real-time inference on edge devices (TensorRT, ONNX, Jetson). Knowledge of web development frameworks such as Angular,Bootstrap,NodeJs
Job Type: Full-time