Have A Big Picture and Envision the End

Three steps to personalize your data science career journey (2)

People without vision perish - Proverbs 29:18

You are passionate about doing?data science and you are on?the journey. However, after you?have worked for a while,?in some way, you are feeling stuck, you feel challenged to move to the next level or you lose your initial excitement. You may not enjoy your work anymore. The above symptoms indicate a missing piece in your data science career. You may never get time to think about where you should end up so you get lost in the rally of data science career.

To keep your momentum and become a champion, you?have?to?see the big picture, understand your purpose and position yourself in the right place. Work is?not about making money or competing for a win; it's?about finding the meaning of doing it?and?how to thrive?on challenges. It's about embracing?problems and turning them into great opportunities?and adding values to the organization you serve and being a blessing to people around.

In this post, I will share with you the data science big picture, how to envision your end and create a unique path to advance your data science career.?Rosabeth?Moss Kanter once said, a?vision is not just a picture of what could be; it is an appeal to our better selves, a call to become something?more.?

Before you can personalize?your path to advance your career, you?have to?know what the key components, functions and related skills are for becoming a successful data scientist, what the highest level you want to reach.

1.?Data Science?Big picture -?key components,?functions,?and?related skills

The Figure below?shows the seven key components or seven phrases of a typical data science project. It starts with business opportunities. Data science aims to solve business problems and create new solutions?or improve existing solutions.?Once you know what problems you want to solve, the second step is to gather all the relevant data followed?by?understanding?the data like quantity, quality and distribution (step 3). The fourth?step is to find the right statistical or machine learning algorithms that fit the specific type?of?problems?best. You may need to do some research?or?experiment before choosing a right methodology. Once you develop a solution, the last two steps?are?to communicate the solutions with business stake holders and build products.?

The Figure below?also shows the main functions or activities involved in each step and related skills required for you to successfully perform the corresponding functions.

For detail information about the life cycle of?a?data science project, please?reference?the two blogs,?Life Cycle of Data Science Projects(1)?and?Life Cycle of Data Science Projects(2)

2.?Envision your end – what is your final position?

The big picture?above,?tells us?that?a?data science project is a team?sport,?and it?requires a group of people with diverse?skills,?and?each has?a right focus. Because?it’s?hard?to find a single unicorn who knows very things and does?all the work.

With?a?high-level?abstraction,?data science?can be projected?into a three-dimension space (Fig 2.?below), business, quantitative foundation, and software development,?that is well aligned with the data science?Venn diagram from Drew Conway’s (2010) and the skills diagram from Grady and Chang (2015)

Based on this, as a data scientist, you may end up in the four possible positions?in your journey:?Analytics consumers,?Analytics solution creators,?Content creators?and?Analytics leaders?or consultants.?The?corresponding?skills?to?the four?end?positions are illustrated in?Fig?3?below.

1)?Analytics consumers – These people have strong business domain knowledge and understand business problems and new opportunities for developing new products or services. They know how to use data and analytics solutions to increase business KPIs. They are great stake holders of analytics solution creators.?Their job title are usually application developers, data driven managers or business analysts, etc.

2)?Analytics solution creators?– these people have strong knowledge in statistics, machine learning and deep learning. They are very good at mathematics and specialized in mathematics and computer science like quantum computing and data mining, etc. If you have strong capability in the area and passionate about finding new solutions by experimenting different algorithms, you can be end in the positions as a principal data scientist or fellow data science researcher to differentiate yourself from others. They focus on discovering new methodologies for solving existing problems or new problems.?

3)?Content creators?– these people are data experts?and understand?database or big data technology, and?analytics platforms. They also have?some?basic analytic capabilities to visualize data?and manage performance.?They are very good at data collection, integration and wrangling etc. They usually?provide?analytics solution creator?with?quality data, help analytics consumers understand data structure and?design?processes, etc. They are data engineers or data modelers.

4)?Analytics leaders?or consultants?– these people have comprehensive capabilities:?understand businesses, machine learning?and AI, data as well as software design and programming. They usually have many years of experience in analytics or data?science?so they have strong business acumen. Given a business problem, they know immediately what type of problem it is from statistical and machine learning perspective, so they are very good consultant on using data science. Those people also have leadership experience. They may start with a technique lead then become a leader in organizations like manager or director or even higher position of analytics and data science.

With step 1 and 2, you should?do self-examine, and recognize where you are right now in your data science career journey. No matter where you are on the journey, you should get some time to figure out and decide?your end position?by leveraging your passion, personality, and strength.

You should start with your passion and personality and nature talent first. It’s better to ask others and see how others think what you are supper good at.?

3.?Craft?your?unique journey

Once you understand the big picture, figure out where you are?and envision your end, then it’s time to draw a road map to follow.?

Suppose they want to become an analytics solution creator, then you should build your fundamental skills in mathematics or statistics field and computer science first.?Later, you can build some business skills that mainly related to different type solutions mathematically.?

If you are?doing?data analysis right now in data?or content?development, but you want to become an analytics solution creator?in the?long term, then it’s time for you to adjust your path so you get more quantitative skills and understand different types of machine learning algorithms and corresponding business problems.?

You can reference career pathway?published?by?Dice,?there are basically 3 levels, entry?level, middle and advance Level.?Individual contributors usually start?as?data analysts?and progress through data?scientist, senior, principal data scientist then reach a leadership position like director, AVP, VP and Chief Analytics officer.

No matter where you?want to end, there is?always a?path?to reach there if you do not give up. During journey, sometimes there is detour you?have to?take; sometimes you have to redraw your path because the nature of dynamics in technologies, environment and life; sometimes?you may need to reconsider your direction; sometimes,?you may unexpectedly?find a better path and?get your destination earlier.?Remember each person will have different path, do not follow others mechanically.

Stories?of advancing in data science career

With?a?doctor degree in mathematics,?Dan?started?his data science career as an individual contributor during?his?first four years and?learned how to solve real world business problems from?four?different companies. Then?he?became a technique lead and learned some leadership skills eventually?he?was promoted to the director of data science.

With two master’s degrees in both mathematics and computer science, I started with as a software engineer. Because of my mathematics background, I did intensive?R&D?in the first?eight?years of my career, where I developed in-depth skills of machine learning algorithms,?search engine?and NLP technology.?The most importantly I developed a mindset of?monitoring?emerging technologies in analytics, data science, big?data?and?distributed computing technologies.?In order to keep my skills most recently, I jumped in a start-up where I learned big data platform and sentiment analysis, and get more familiar about social media?contents,?also?I?developed a keen?mindset to identify new business opportunities and develop new products?or services. Later, I realized I could add more value to a business if I could mentor or coach others. So, I made another move in career to become a manager then a director of data science right now.?

During the journey, there are always new challenges to conquer. But I think any challenge is a new opportunity to?find untapped potentials;?and any problem is a new opportunity?to develop?a?new?solution. Sometimes?we need to take?risks?and build?our new capabilities and faith spiritually.?People tend to give up easily?when facing?obstacles, but?perseverance?is key to make your dream come true.?Embrace a positive mindset to tackle through?any?obstacle. Once you?climb the mountain, you reach to a new?height?then be ready for climbing a?higher summit.

In the next post, I will write Step 2, Build the data science foundations.

Stay tuned if you are interested.