AI HelpDesk Dev (Python)

PROJECT OVERVIEW

Current capabilities of the help desk and manual advanced troubleshooting capability lack the tools to meet customer service quality expectations. There should be a systematic way for troubleshooting thousands of lines of logs data.

The goal is to make the process as seamless, automated and knowledge-driven as possible to resolve issues quickly while continuously improving product quality and Increase capacity to find root causes. The AI-enabled workflow introduces a more efficient and automated process by querying thousands of tickets, multimodal data pre-processing and analysis. This speeds up the process and scales it to handle a much larger volume of tickets. This helps a significant leap in helpdesk efficiency, speed, and scalability.

Data Collection, Data Pre-preprocessing

Implement NLP techniques and leverage domain expertise to process complex data effectively. Development of modular, configurable Python code for feature extraction, Exploratory Data Analysis (EDA), performing clustering, and analyzing patterns to identify and engineer relevant features for different target types.

Pipeline creation for offline data analysis

Creating a configurable pipeline for data preprocessing, model training, and evaluation to accommodate different root cause models. Integrate feedback from outliers to ensure ongoing model improvements, making the system adaptable and self-enhancing over time.

Model enhancements for different target type root causes

Develop an ensemble of models for 20+ root causes across various target types. Rigorous feature selection, model development (Supervised ML), testing, and validation. Fine-tuning models to improve accuracy and generalization, ensuring robust performance across a wide range of scenarios.

Model integration, deployment and MLOPs

Design & implementation of a Continuous Deployment (CD) pipeline, integrating logging services for system monitoring & debugging, monitoring tools for real-time performance tracking. Online pipeline testing and validation procedures to ensure the systems reliability. Achieve seamless integration of models into the production ecosystem, maintaining high performance and reliability.

Required skillset - The ideal candidate will have a strong background in Python programming and experience with machine learning and natural language processing tasks. This role requires a blend of technical expertise and practical experience in developing and deploying AI solutions.

• Minimum of 2 years of professional experience in Python development

• Proficiency in software development tools and practices, including Git, DVC, unit testing, code coverage tools, and Docker

• Experience with web frameworks such as Django or Flask

• Strong background in Natural Language Processing (NLP), including:

  • Regular expression implementation

  • Data engineering and cleaning techniques

  • Feature extraction methodologies

• Basic Machine learning, clustering models knowledge and experience.

• Demonstrated ability to work with large datasets and complex models

• Experience in logging, troubleshooting, and optimizing ML models

• Familiarity with software testing practices and performance optimization

• DevOps experience, particularly with Amazon EKS (Elastic Kubernetes Service)

• Experience with MLOps and model deployment on cloud.

The candidate will be responsible for developing, testing, and maintaining AI/ML solutions, collaborating with cross-functional teams.