Ansible Build Time Prediction

Project Description

Summary

We would like to build a continuous learning algorithm that will be able to predict execution times of Ansible builds (Playbooks) based on historical Ansible build data.  In a complimentary project we are generating Ansible build data from which the algorithm can learn.  The winner of this project will be responsible for the creation of the continuous learning algorithm and API for making Ansible build predictions.

As part of your proposal please answer the following questions:

•What kind of machine learning algorithms would you use to solve this problem?

•What trade-offs you are making when choosing one algorithm over the other?

•Which technology stack would you use for this challenge?

•What are the underlying assumptions about the training data set?

•How would you approach in tuning the parameters for the chosen algorithm?

•How do you plan to evaluate the performance of the trained machine learning Model?

•How do you plan to develop the API?

Scope of Work

The selected consultant will be responsible for:

•Selecting the machine learning algorithm which optimally predicts execution times of Ansible builds

•Training the model on the historical data of the Ansible builds

•Building an API for evaluating the performance of the model and tuning it to improve further

•Executing the API on the actual test data

•Creating a feedback loop which provides a continuous machine learning environment

The primary output of this project is an API for predicting Ansible build times with the implementation consisting of continuous machine learning algorithm which ingest a data source of Ansible build factors.

The attached presentation provides additional details around the environment and machine learning and gives additional context to the broader project scope (including the data generation project).  Details completed by previous projects or otherwise out of scope for this Experfy project posting have been greyed out for scope clarity, however the details may still be relevant to your implementation.

Challenge Format

We plan to hire more than one expert to implement their model using a common initial data set.  The different approaches will be evaluated after initial implementation and one of the two experts will be asked to continue with the project refining their model and build the API.  The period for determining which approach will be used (and who will complete the final project deliverable) will be variable but is expected to last 1-2 weeks.  For your proposal, please specify a fixed cost for delivering the entire project.  Prior to hire and starting the project, we will negotiate partial payment (whether hourly or fixed) guaranteed to both experts upone completion of the initial model, regardless of the expert being selected to complete the project for full payment.