Project management is a critical process for the success of the project. Big data, ML, and Business Intelligence projects have different priorities and challenges.

Even simple machine learning projects can be complex and managing these projects in a real business is much harder than most people realize; that’s why VentureBeat claims 87% of machine learning products never make it into production, and Harvard Business Review says that “The first wave of corporate AI is bound to fail.”

FOCUS ON PRODUCT NOT ON AI

AI is hype and when you use “AI-powered xxxx” people are more interested in your project. The popularity of AI causes to people develops unnecessary products.

Product is more important than AI. Understand your customers’ needs and develop the easiest solution for them.

Deterministic to the probabilistic approach

AI systems differ from traditional software in many ways, but the biggest difference is that machine learning shifts engineering from a deterministic process to a probabilistic one.

With AI, you often don’t know what’s going to happen until you try it. It isn’t uncommon to spend weeks or even months before you find something that works and improves model accuracy from 70% to 74%.

The main difference between classical programming and ML is inputs and outputs. So, you should think about which approach is more applicable to your problem.

Is it possible to build an MVP without ML?

Don’t be afraid to launch a product without machine learning.

Machine learning needs data and the result will be probabilistic. So, in the MVP stage, starting with small size data and low accuracy can create unsatisfied users.

My advice starts with rule-based solutions and step by step develop ML in the background. When you feel that ML results are better than traditional rule-based systems you can transform the solution into ML. On the other hand, to reach more data you need a working and attractive MVP. So, try to use more mature technologies to take results.

If your project is in a group described below, do not start with ML.

My advice

I have worked on several companies’ ML projects. My suggestions:

1- Do experiment: You use different ML models to solve a problem. Hence you should make experiments. Unfortunately, sometimes you can’t reach an acceptable result. So, in AI-powered projects, you should make a detailed risk mitigation plan, especially for data and ML problems.

2- Focus on customer needs: Don’t be obsessive about ML usage. Be practical and develop the most useful solution. Sometimes this can be a simple function instead of complex ML works.

3- Ask for advice from ML Experts: Take consultancy in the project planning stage. Sometimes you can create elevated expectations about AI implementation. If you don’t aware of the limits of technology, you can live problems in project management. Please remember theory and practice is really different in ML projects.

4- Use Cloud Infrastructure for ML model development: Cloud is a big advantage for AI projects. You can use prepopulated ML models, computational power, and data pipeline without the need for maintenance. So, select a cloud provider to implement your data project.

5- Work on Data, digest your data and make a data management plan: ML projects start with data. If you don’t have a data pipeline and business intelligence projects, it won’t be easy to develop a unique ML powered solution.

6- Don’t use ML without an appropriate plan: In ML projects make a clear plan about your workflow. Otherwise, your project starts to be an experimental work and you can’t manage your resources.

7- Don’t launch with insufficient data: To complete on time, don launch your product with small data set. Then your ML model makes drastic errors, and your project loses its reliability

8- Use ML to help with data collection: Data collection will be your biggest problem in ML projects. So, use predefined ML models to better data collection. Please search for ML solutions like image tagging, missing value forecast in a data set, Natural Language Processing and anomaly detection to enrich your data set.

Botmore Technology

We are supporting companies in ML project plans with really affordable prices. Please send an email to [email protected] for more details.