30,000 students and growing!

This week my total number of online students reached 30,000 and I wanted to share an important lesson with you about what I’ve learned from teaching others.

Throughout my 18 years providing technology-based solutions to startups and corporate, I’ve always had a desire to teach others what I’ve learned. That’s why I began to develop online courses based on my expertise in Leadership, Software Development, and DevOps with the goal of educating others to help them become passionate, highly competent software engineers. It’s two years since I started teaching online and I’m grateful to be currently teaching over 30,000 students around the world through LinkedIn Learning, Lynda and Udemy. It gives me great joy to read reviews from my students and hear how my teachings have positively impacted and helped them to progress.

The lesson I’ve learned is that we have a responsibility to share our knowledge and experience to help others following behind us and that teaching other is truly fulfilling. Whether we teach one person or thousands of people, having a beneficial impact on others, teaching them new skills, boosting their confidence and seeing them progress in their career is priceless.

Do you feel the same sense of fulfillment when you help others?


Proof that More People Solve Problems Faster and More Efficiently

In Leadership, it is proven by experience and by common sense that, if you have (for example) 50 problems to solve for a team of (again for example) 10 people in a given time (say 4 months) the team can solve the 50 problems much faster and more efficiently compared to when you give 5 problems to each member of the team (and each member works on their own 5 issues only).

For those who need numbers to believe this, here is the proof:


Let X be the number of problems,

n be the total number of the team and

t be the time taken to complete a task by an employee

We wish to show that a can be accomplished faster and more efficiently compared to when you divide the task equally among each employee;

  • 50>10*(5) where 50 is the number of tasks to be accomplished, and 10 is the number of employees. 5 is the number of tasks divided within each employee; 50/10

For the first case, the task to be done = x

Assuming that t is directly proportional to x

for the second case, the task to be done = x/n +x/n + x/n+….+x/n =

Using induction method, we wish to proof that >n

Take x/n =I,    


For all fixed nN,

For 1 step summation of 1,


Solving by induction process, . Now consider

= +n+1


This therefore is a proof that by induction, the theorem holds.

We recall our i=x/n


For 1 step, = 1(2)/2 =1

For 2 = 2(3)/2 = 3, for 3=3(4)/2 =6

Say n =10 and x=50,

Then x/n =5

= 30/2

=15 which is greater than 5.

Recall that t is directly proportional to x; therefore, it will be faster for the problems to be solved by a team within a given time period compared to dividing the problems within the group.

Aref Karimi


Why are individual goals/KPIs outdated, and how do modern organizations use KPIs to thrive their teams?

Google’s new cross-industry survey about key performance indicators (KPIs)/Goals, shed light on the challenges and emerging opportunities companies face when using KPIs, demonstrating the many ways advanced use of KPIs can benefit organizations, and offer steps leaders can take to make the most of KPIs going forward.

Old school managers (who tend to lead their organizations based on books written in the 1950s), use KPIs (or Goals) to (micro) manage their organization and measure the performance of each individual based on their past. These KPIs are known as Measurement Capable and Measurement Challenged.

Use KPIs to lead teams rather than to manage individuals

Let’s quickly have a look at the obvious downside of when KPIs are set for individual team members rather than for the teams. Then we will explore how modern organizations use KPIs to lead rather than just to manage.

The first obvious downside of an individualized KPIs is that the team spirit will die out and a given team will fail to operate as a unit with a shared goal. This becomes even a larger issue when KPIs or goals are bound to bonuses and remunerations, as each person will focus on ticking their own goal off in order not to sabotage their bonus, rather than helping the team to thrive and perform at its best.

For those who need numbers rather than common sense only, I have done a mathematical calculation which proves my point!

So now that we have the proof that team KPIs are way more efficient than individual goals, it is worth noting that rather than focusing exclusively on how KPIs can help them manage their organization, leading companies look to KPIs to help them lead — to find new growth opportunities for their company and new ways to motivate and inspire their teams. Simply put, KPIs will be used as a guide for a better future rather than for the past.

Use KPIs to align the organization

Now that we have proven the obsoleteness of individual goals, we can see how the team KPIs can help align the different teams or departments effectively because an organization cannot succeed if they isolate internal, employee-related KPIs and financial or process KPIs from the customer experiences (they endeavor to create). Leading organizations use KPIs to effectively align people and processes to serve the customer and in general the company vision.

But how team KPIs can be used for organization alignments? How do we know what the client KPIs can or should be?

To answer this question we need to remind ourselves that KPIs are a tool for leaders to set the right vision and come up with correct strategies for their future. Thus, in order to use KPIs as a navigator rather than a performance management tool, KPIs must be turned into insight, or simply put, into data!

One major data that can help us, leaders, to come up with efficient KPIs that can serve our customer and our brand is customer feedback. If the customer experience is recorded and processed (by data analysis experts) then it can offer us invaluable insights which can be used by product managers, marketing leaders, sales directors as well as by the executives.

Although the fast pace and smaller companies must be careful not to slow themselves down by waiting for insights before making every decision, we must look at insights as our guide to set the strategies right.

Use KPIs for machine learning

Now that we live in the machine learning era and artificial intelligence is booming, ML/AI technologies can help us achieve enterprise goals.

The greater potential of ML is empowering software and systems to learn from data-driven experience. This creates opportunities to use KPIs and their underlying data to “train” ML algorithms. That is, KPIs can be used, individually and collectively, to teach AI systems to improve and optimize their performance.

KPIs are not the target, they are tools

There is no magic number for the optimum number of KPIs for an organization or a team. While a properly set number of KPIs could compel the managers to bring a laser focus to the team’s most essential strategic goals, too many KPIs easily become unwieldy, unmanageable, and create unrealistic expectations; too few might result in the neglect of critical business issues.

From my experience, the best way of coming up with the right number and reasonable complexity of KPIs, is by both learning from the past and by adhering to SMART goals.

As we learned before, KPIs are for us, leaders, to learn from the past. So if by looking at past data we learn that the team failed to deliver their goals in time or with the desired quality, then the KPIs must be reviewed.

One common issue with the KPIs and Goals, especially when there are too many of them and when they are set for each individual person in the company, is that goals deviate from being SMART. SMART goals are Specific, Measurable, Attainable, Relevant and Time-bound. When companies set goals for each individual team player, making the KPIs specific becomes hard and the expected outcome becomes vague. Even more so, in a fast-paced organization, since things change frequently, goals tend to lose their relevance and attainability. And this sets the teams and people to failure.

What do we learn?

We learn that individual goals and/or KPIs are frowned upon in modern leadership. Teams can deliver way more efficiently if KPIs are set for the team rather than for the individuals.

We also learn that KPIs must not be for measuring the performance and should not be bound to remuneration and/or bonuses. At least not only for that. Rather KPIs must be used for better future strategies and for better alignment of various teams and departments in the organization.


TeamCity vs. Atlassian Bamboo


A comparison between TeamCity 2018 and Atlassian Bamboo!

One of the things I always do for the software development teams that I lead is introducing them to CI/CD, and setting up an effective Continuous Integration and Delivery system for them.

Often when it comes to implementing a CI/CD system there is a debate as to what CI/CD tool is the best to use? Some are fond of TeamCity while the others are a fan of Bamboo! And of course Jenkins has always had its lovers!

Since I have published two courses about TeamCity 2018 and Atlassian Bamboo, I would like to compare those two here and see which one would suit you better if you were to setup your CI/CD system.

When comparing tools such as TeamCity and Bamboo, we need to look at all aspects of running them rather than just looking at their features. In this article I will compare them in regards to their features, support, price and so forth.

TeamCity 2018:

TeamCity 2018 is NOT Open Source so if you use Open Source products you better think of a tool such as Jenkins or Sand Castle. To use TeamCity you need to buy a proper licence.

In terms of East Of Setup, TeamCity will get the TOP WINNER’s award amongst all other CI/CD tools. Installation of TeamCity, especially on Windows, is so straightforward and easy. Even installing TeamCity in a Highly Available and Highly Scalable manner is not difficult.

TeamCity also has the biggest number of Built-in Features amongst other CI/CD tools in the market! You can configure builds, define templates, define build triggers, configure dependencies etc.

TeamCity executes the build steps sequentially and it won’t run your build steps (e.g. compile and unit test) in parallel. This may not be ideal if you need to build, test and package big projects fast.

In this day and age many companies put their infrastructure on the cloud. If you use AWS, the good news is that TeamCity supports Cloud Build Agents on Amazon Web Services. This means that you will create a template for your Build Agents and TeamCity will launch the Build Agents based on those templates when needed.

Now let’s talk money!

If you have a small team which needs no more than three build agents, and won’t need support from JetBrains, then you can a Professional Server Licence, which is Free! With this licence you can have up to 100 Build Configurations, which for many small teams is more than enough!

If you need more build agents, you can pay $299 for each additional Build Agent and get Build Agent licence.

Bigger companies who need more build configurations, as well as official support from JetBrains, would have to buy an Etnerprise Licence which costs between $1,999 and $21,999!



Atlassian Bamboo:

Bamboo is NOT open source! You will get a Java package that can be virtually installed on any machine! The good thing is that since Bamboo is a Java application you will deal with the same application on different operating systems.

It is easy to install Bamboo especially on Windows, because as said before it is basically a Java application. TeamCity is a tad easier to setup comparing to Atlassian Bamboo especially when it comes to connecting to the database because Bamboo uses JDBC drivers which may be a bit unusual for most DevOps engineers and developers.

In terms of Built-in features Bamboo is behind TeamCity for sure! However the structure of your projects in Bamboo can be a lot more flexible! In TeamCity you only have “Build Steps” that execute sequentially, but in Bamboo you will have Projects, Steps, Jobs, and Tasks! This will give you a great deal of flexibility especially that Jobs can execute in parallel! That means while you run your unit tests you can run code inspection too! Or do many other things that can happen in Parallel, and save time!

Bamboo does support Cloud Build Agents aka Elastic Build Agents. However at the time of writing this blog post, Bamboo cannot be installed for High Availability and High Scalability.  You will need to install one instance of a Bamboo server and run a second instance as a backup! Nicht gut!

Bamboo has a diverse pricing options which you can see HERE. If you are a solo developer or a very small team you may use Local build agents and pay as little as $10. For more sophisticated setups you will pay something between $1100 to $126,500!

You can use both Atlassian Bamboo and TeamCity to build, package and deploy many different types of applications written in many different languages. However Bamboo is more favoured by Java developers whereas TeamCity is more favoured by .NET developers. Maybe because TeamCity comes with full built-in support for MSBuild and PowerShell! In saying that, virtually there is not any program that you cannot compile with both of these tools as long as there is a compiler for that program!

Can Techies be Leaders?

Although I am a techie myself one thing that I am passionate about is teaching management skills to new managers and talents both online and on-site.

One thing that I have learned myself through teaching management and mentoring talents is that transitioning from a tech person to a manager is quite difficult for most people. I often hear from my senior engineers that “I do not see management capabilities in myself”.  You may think that this may be a self esteem-related issue, which is somewhat correct, however we have to bear in mind that lack of confidence can rise from lack of knowledge and skills.

I have also witnessed that companies tend to promote technical staff (e.g. senior developers) to tech leadership positions. Although in rare cases it turns to be a successful experience, in most cases it does not end well.

Learn How To Become A Manager Now!

In fact my personal experience with this has been appalling! In 2005 when I was asked to manage a team for the first time, I was actually a senior developer who had many good ideas and was able to deliver work fast and with good quality. But I am not sure if it was enough for me to become a manager. From what I remember, the first few weeks went well but then issues started coming up. There were quite a few reasons for that and amongst them the most important one was my “Technical Mentality”!  A “Technical Mentality” is the thinking process of technical guys (e.g. engineers, developers etc) which is very black and white. 1 + 1 is always equal to 2 in a techie’s mind, and white is always white!  Plus, they tend to think that they are always right, and it is too hard to convince them otherwise!

So you may ask that “isn’t 1+1 equal to 2?”. Or you may ask “What do you mean that white is not always white?”.

The fact is that for good managers and for leaders, there is always a gray area and they tend to find a win-win solution when the disagreements arise. They are flexible and instead of insisting on their own opinion or solution, they listen actively and try to underestand where the other person comes from. For them a win-win situation is better than a triumph in a technical debate which  may damage the other person or even the entire team’s engagement, positive energy and moral.

You may not believe that how many tech guys I know that they dislike someone else just because the other person is not as good as they are in technical subjects! Unlike these types of techies, leaders never like or dislike any of their teammates for their strengths or weaknesses. A manager’s job is to focus on people’s strengths and help to improve their weaknesses. What I believe and  always say is that: If you do not love your team, you cannot do anything good for them!

But is this “techie mindset” fixable? My answer to this questions is “Hell yes!”.  A techie mindset is like a crooked tooth! What it needs is some force in the right direction and then over time it will go where it should go to! The force for tech people and new managers is courses and/or books, and direction comes from mentoring! Just like any other job, one can become a manager by learning the concepts and skills, then putting them in practice and making them a habit!

Setting up a TeamCity 2017 Cluster

It is very common for companies, teams and engineers to setup TeamCity in a way that TeamCity will not be able to unleash its real power!

The main two major .mistakes that are made in setting up a TeamCity-based continuous integration and delivery system are:

  • Putting the build agent software on TeamCity server (server hosting both TeamCity software and the build agent software).
  • Installing TeamCity as a single server.

Although TeamCity allows you to to install both TeamCity web application and its build agent service on the same computer it does not mean that it is a good idea! Only for learning purposes can someone install both TeamCity software and TeamCity build agent on the same machine otherwise this setup is not good and will cause headaches. So why should not you put both the TeamCity 2017 web site and its build agent on the same computer?

  • Because servers must be single responsible! One server should not have two responsibilities, which means your server must be either a TeamCity server or a Build Agent server.
  • Because build agents run the risk of getting corrupt and you may need to kill them at times! If this happen then you will have to kill your TeamCity server too!
  • Because you may want to take advantage of cloud build agents, in order to save $$$. If you set up your build agent on your server, you will not be able to terminate/stop your build agents when you do not use them.
  • You cannot scale out your build agents. The more concurrent builds you have the more build agents will be spun up by TeamCity but if you put the agent and the web-app on the same server this cannot be done.

Also setting up TeamCity as a single server does not seem to be a good idea for being used in production capacity as it will not be able to scale out. The figure below shows a typical (and not ideal) TeamCity setup:

SIngle-server setup of TeamCity

As you see the TeamCity server will be able to spin up multiple TeamCity Build Agents (if you setup cloud build agents) but it will not be able to scale out the actual TeamCity application. It will also not be highly available as a failure in the TeamCity Server will take down your CI/CD system.

In order to have a highly available and highly scalable TeamCity setup it is needed to use a centralised database such as SQL Server or MySQL so that all the projects, build configurations, templates etc can be shared amongst TeamCity servers.

Apart from a shared database, the data directory of the TeamCity servers have to be shared too so when you install the servers make sure you choose the same shared (e.g. network) location for all servers. You can also update the path of the data directory later via modifying the configuration file.

Copy of LearnTeamCity.pngScalable and Highly Available setup of TeamCity

Speaking in Amazon Web Services language, in the above diagram, our TeamCity Serves are part of an Auto Scaling group, which means that as stress and load goes up on the servers, AWS will spin up more and more TeamCity servers.

If you want your set up be highly available as well you will have to make sure that your Auto Scaling group stretches across multiple availability zones (physical locations) too.

To learn more about CI/CD with TeamCity 2017 and AWS CodeDeploy, you can see this online course. I have set a very special price on this course for my blog readers 🙂

Click to get 85% discounted online course about CI/CD with TeamCity 2017 and AWS CodeDeploy

TeamCity 2017: Build and deploy the modern way!

Grafana, Graphite and StatsD:Visualize your metrics!

Hey guys,

I have recently published two new courses on Udemy, and one of them has become a “Best Seller” so fast, in only two weeks! This course is interesting for a lot of DevOps engineers because monitoring and visualising the metrics of infrastructure, websites and applications is an absolute must-have skill for all DevOps engineers!

Developers also show interest in this course because the almighty Grafana-Graphite pair is an excellent tool for instrumentation and health check of application and websites.

In my course I have explained in over 25 lectures and 5.5 hours of videos that how you can extract and visual metrics from various sources whether it is supported by Grafana out of the box or not!

If you are an IT professional I highly recommend this course to you because based on my experience good instrumentation and having visibility on your systems metrics is what differs a good software system from the average ones!

Here is a coupon for my blog readers, which offers 80% discount!

Grafana and Graphite for DevOps Monitoring