MUST HAVE:
-Recommendation Systems/Recommendation Models Expertise
-Matrix Factorization/Econometrics experience
Responsibilities-
- Design and develop machine learning algorithms and deep learning applications for recommendation models.
- Fine-tuning open-source models and deploying them for real-world applications.
- Experience with OpenAI models (e.g., GPT-4), creating custom AI solutions to answer queries, assist sales teams, and provide recommendations.
- Solve complex problems with multilayered data sets and optimize existing machine learning libraries and frameworks.
- Ensure ML algorithms generate accurate user recommendations.
- Expertise in implementing hybrid search and retrieval-augmented generation (RAG) techniques.
- Familiarity with Matrix factorization, collaborative filtering and ranking.
- Train, retrain and monitor machine learning systems and models as needed.
- Development of high-performance search applications, including product search and keyword-based search using tools like ElasticSearch and OpenSearch.
- Advanced testing and evaluation of different chunking strategies for optimized performance.
- Hands-on experience with building and managing large-scale machine learning systems.
- Deep knowledge of infrastructure-side challenges, such as scaling models, load testing, and ensuring high availability.
- Proficient in AWS services, including Bedrock, Redshift, Airflow, and Databricks for data and model pipelines.
- Strong focus on performance optimization, continuous integration, and improving ML systems for deployment at scale.
- Extensive experience leading machine learning projects end-to-end, from design and development to deployment and monitoring.
- Collaborates closely with stakeholders, ML engineers, data scientists, and DevOps teams to ensure successful project delivery.
- Builds out evaluation frameworks that incorporate user feedback from logging and fine-tuning model performance accordingly.
- Building robust data pipelines for machine learning models, ensuring that data is clean, properly preprocessed, and available for model training and deployment.
- Expertise in automating ML pipelines using Airflow and optimizing workflows in distributed environments.
- Experience in integrating and managing large datasets for training complex models, including deep learning frameworks.
Technologies and Tools:
- AI/ML Frameworks: Personalization & Search Systems
- Search Technologies: ElasticSearch, OpenSearch, Hybrid Search, Keyword Search
- Cloud Platforms: AWS (Bedrock, Redshift, S3, EC2, etc.)
- Data and Workflow Tools: Databricks, Apache Airflow
- Programming Languages: Python & Java (With libraries such as scikit-learn, TensorFlow, PyTorch, XGBoost, Hugging Face, etc)
- Other Tools: Model performance evaluation frameworks, logging and monitoring tools
