According to a report, data scientists spend two hours a week searching for new jobs

‘I’m a data scientist,’ feels pretty prestigious to say this, isn’t it? Then why is there a downward trend recently in data science professions, and especially, among data scientists? Lately, for over a couple of years, data scientists are quitting their jobs from top technology companies. Despite getting paid handsomely, they choose to walk out on many scenarios. The worst case is that most of them don’t even complete a whole year in the company.

…Data scientist was named as the ‘sexiest job of the 21st century’ by Harvard Business Review not long back. Starting from Fortune 500 companies to retail stores, organizations around the world want to build a team of top data science professionals to drive their company towards success.

Despite getting a lot of attention for a long time, the positive trend is taking a u-turn in recent years. According to Financial Time’s investigation, data scientists are spending an average of two hours a week looking for a new job. While machine learning specialists topped the list of developers who said they were looking for a new job, at 14.3%, data scientists followed the trend with 13.2%. Read More

— Some of the Best Places to Explore — A curated list of dataset repositories for deep learning projects and good places to look for additional benchmark datasets for your model, so I am putting it out here, hoping you find it useful! Read More

#data-science

…This is a long post divided the post into 6 categories. Feel free to read categories that relate best to your role as a data engineer, data scientist, ML engineer, data-business leader:

• ML Problem definition: The formative stage of defining the scope, value definition, timelines, governance, resources associated with the deliverable.
• Dataset Selection: This stage can take a few hours or a few months depending on the overall data platform maturity and hygiene. Data is the lifeblood of ML, so getting the right and reliable datasets is supercritical.
• Data Preparation: Real-world data is messy. Understanding data properties and preparing properly can save endless hours down the line in debugging.
• ML Model Design: This phase involved feature selection, decomposing the problem, and formulating the right model algorithms.
• Model Training: Building the model, evaluating with the hold-out examples, and online experimentation.
• Operationalize in Production: This is the post-deployment phase involving observability of the model and ML pipelines, refresh of the model with new data, and tracking success metrics in the context of the original problem.

Read More

Data Scientists aren’t born — they’re made. IT pros from all backgrounds are working to gain the types of all skills companies need as the demand for data scientists outspaces the supply of qualified candidates. These are some common personality traits and skills of a data scientist. Read More

#data-science