AI Search Engineer

Overview

We are seeking an AI Search Engineer with a deep understanding of large-scale information retrieval systems, deep learning, databases, and Retrieval-Augmented Generation (RAG) architectures. The ideal candidate will have expertise in developing and optimizing search algorithms, implementing efficient indexing techniques, and leveraging RAG to enhance AI-powered search and question-answering systems.

Key Responsibilities

RAG System Research and Implementation: Lead the design and implementation of advanced retrieval systems like Deep Memory by Activeloop, delivering optimized RAG systems across the entire value chain - from embedding or model fine-tuning to retrieval optimization with custom algorithms, to enhance knowledge retrieval accuracy.
Search Algorithm Optimization: Develop and refine search algorithms, including semantic search, hybrid search, and multi-modal search techniques, to improve retrieval performance and relevance ranking.
Vector Database Integration: Implement and optimize vector storage and indexing solutions within Deep Lake, ensuring efficient similarity search capabilities for high-dimensional embeddings used in RAG systems.
Query Understanding and Processing: Design and implement advanced query processing pipelines, including query expansion, intent recognition, and contextual interpretation to enhance search precision.
Information Retrieval Model Development: Create and fine-tune machine learning models specifically for information retrieval tasks, such as document ranking, query-document relevance scoring, and zero-shot retrieval.
Performance Evaluation and Metrics: Establish comprehensive evaluation frameworks for search and RAG systems, including relevance assessments, A/B testing, and user satisfaction metrics to continually improve system performance.
Scalability and Efficiency: Optimize RAG and search systems for high throughput and low latency, ensuring they can handle large-scale datasets and real-time query processing demands.
Data Ingestion and Indexing: Develop efficient data ingestion pipelines and indexing strategies to support rapid updates and real-time search capabilities across diverse data types and sources.

Qualifications

Master's or PhD degree in Computer Science, Machine Learning, Statistics, or a related field.
Strong programming skills in one or more programming languages, such as Python or C++.
Extensive experience with machine learning libraries, such as TensorFlow, PyTorch, Llama Index, LangChain, etc.
Proven experience in developing and deploying complex machine learning models in production environments, including experience with cloud-based platforms, edge devices, or embedded systems.
Strong understanding of advanced machine learning algorithms, such as deep learning, reinforcement learning, RAGs, and ensemble methods.
Experience with model optimization techniques, such as hyper-parameter tuning, model compression, and quantization.
Solid understanding of data pre-processing, feature engineering, and data quality assurance techniques.
Experience with large-scale, complex data sets.
Excellent problem-solving skills and ability to analyze and interpret complex data sets to extract meaningful insights and drive decision-making.

Preferred Qualifications

Experience in training deep learning models in a distributed manner.
Publications in top-tier machine learning and AI conferences such as ICML, NeurIPS, and CVPR.
A builder attitude and a passion for developing hyper-scalable software for ML.
Ability to proactively identify and anticipate problems and provide tangible solutions.
Enthusiasm for the startup journey of building an endurable, scalable business.

About Us

Activeloop is a cutting-edge company focused on developing Deep Lake, a database for AI powered by a unique storage format optimized for deep-learning and Large Language Model (LLM) based applications. We simplify the deployment of enterprise-grade LLM-based products by offering storage for all data types, querying and vector search, data streaming while training models at scale, data versioning and lineage for all workloads, and integrations with popular tools such as LangChain, LlamaIndex, Weights & Biases, and many more. Join us in our mission to revolutionize AI data management.