I am looking for a talented full-stack developer with expertise in machine learning for a project which is fairly simple in concept.

The goal is to implement a machine learning system that is capable of scanning a set of documents and using the data to do the following:

• Create a digital baseline of  a documents and automatically compare it to other revisions added to the same file location.

• Use ML and training to sort the data highlighting the differences  and train to provide sensical data.  For example, training the system to ignore dates, document numbers, generic text etc to only show meaningful changes.

The second phase of the ML training is based on:

• Extracting coordinates from the data sets and using them to group and sort documents.

• Train system to automatically scan, sort and assign documents to matching coordinate / tiles linked to a UI.   

The above will be done  for each data set, normally twice a per year, however it should be automated to carry out the tasks as soon as new data is added to that specific file.  

At present there are 400 minor data sets, which are approximately 10mb each, which will need to be scanned and baselined and then ML applied.

I will need advice if both phases are achievable, if so, how what is the approximate timeline and cost to get to MVP. This can be a rough figure initially, which we can refine as the project progresses.

Project Stage:

The project is currently in the feasibility stage and the decision to proceed will be made early next week. The product is well positioned, with a target customer, direct need for the solution and existing budget allocation.

This project will need to move forward fast and efficiently if  given the green light.   

Budget:

This project is currently operating on a shoestring budget, as it is in its utmost infancy.

The solution has been formulated from insider knowledge to a specific engineering problem, and requires an immediate solution.

There are critical contract renewal dates to consider for current suppliers that dictate a quick start on this project.  

I am happy to be  take a slightly scrappy approach to get to MVP.

There is no estimated budget for this product as I’ll firstly need to get software development costs and review if it will be worthwhile.   

The intention is keeping the cost as low as reasonably possible to get to the MVP, which will be enough to secure a yearly contract, which will fund continued product development.  

Who?

Someone who understands the efficient flow of end to end MVP development and has an appreciation efficient front and backend coordination.  

Extensive experience in ML isn’t essential, however a genuine interest in the subject is preferred.

Please only quote if you have done something similar before.

Many Thanks

Posted On: January 20, 2024 15:00 UTC
Category: Full Stack Development
Skills:Web Application, Machine Learning, UI/UX Prototyping

Country: United Kingdom

click to apply