How you can create a strong AI talent development strategy

Artificial intelligence (AI) is revolutionizing our way of life by automating decisions, predicting outcomes, and optimizing processes. From our phones to shopping, medication, banking and manufacturing, AI is everywhere.

However, there is growing concern that advances in AI are being slowed down by a shortage of trained talent that’s needed to scale AI solutions across organizations. This talent shortage is slated to cause a massive imbalance in AI adoption and its scalability across the enterprise.

But what is causing this shortage of talent? Is there really a shortage, or is the problem our inability to utilize talent effectively?

There is much discussion across forums about the right enablement and talent strategy for AI. But the underlying problem is not the lack of skills but the lack of the right individuals connecting with the right opportunities. There are many extraordinary people in the market who would be perfect fits for a career in AI, but the industry simply is not doing enough to provide the right platform to launch their careers.

That’s because there are no best practices and standards developed for the next generation of deep learning and AI skills, and adoption at most organizations is still nascent. Even several entrenched players do not have a strong talent development strategy in place for nurturing their existing AI/ML talent.

An AI talent development strategy

The solution lies in creating a strong talent development strategy along with the right platforms and frameworks for talent to be cultivated, by:

1. Identifying those best fit for enablement programs: From backgrounds like mathematics, statistics, computer science and economics we can get a talent pool that is already acclimatized to structural problem-solving. Similarly, there are people with experience as data engineers, data scientists and machine learning (ML) experts who can be coached and mentored into AI roles with very little transition time. A proper filtration mechanism that selects candidates with the right aptitude and learning potential is key to solving the skills gap problem.
2. Enabling career transitions: Apart from identifying the most suitable talent, there must be well-designed enablement programs to equip talent with the right skill sets. These enablement programs can take the form of bridge programs of short duration, or fully comprehensive training of six to eight months. Apart from that, creating customized growth plans that take aspirants closer to their desired career profile step by step will be another vital ingredient for the transition process.
3. Building robust best-in-class in-house learning platforms: Developing learning platforms for upskilling and reskilling in niche areas is vital. These need to be learner-friendly and provide engaging content and a wide variety of resources and content to enrich the talent pool. These portals can be monitored through analytics. Personalized guidance can be offered to users for better engagement and better learning outcomes.
4. Nurturing partnerships with startups, MOOC platforms: Companies need to invest in partnerships and training for employees with open-source experience and startups specializing in various AI domains. Through partnerships, two-way knowledge transfer is initiated, with mutual enrichment of talent a natural outcome.
5. Nurturing partnerships with universities and think tanks: Collaboration with academia, universities and research organizations, AI consortiums and think tanks brings access to state-of-the-art training materials and research. Academia can also leverage industry feedback to tailor their courses to specific business needs.
6. Initiating mentoring programs from experienced AI professionals: Engaging experienced professionals who can provide the much-needed support and knowledge to train the rest of the team is vital for disseminating the much-needed added skills and technical know-how. Equipping and designating trainers from within the team will cause faster learning and foster a learning culture within the team.
7. Creating incentives: Focus on creating a proper incentive structure to nudge employees toward continuous upskilling.
8. Sponsoring temporary gig projects and job rotation: Creating a support system for employees to work on side projects and hobby projects within the framework of their organization, as well as rotating job roles at proper intervals, is another strategy that can help bolster the skills and provide a better platform for talent development.
9. Instituting hackathons and Ideathons: Hackathons are one of the best ways to get the talent pool hooked into cutting-edge technologies and to give them valuable knowledge. Employees participating in AI hackathons for knowledge-building can see what AI is all about and may become intrigued and want to get more involved.
10. Creating a steady pipeline of entry-level talent: There are very few entry-level positions available in AI, which makes it hard to develop fresh talent. Many times, the recruitment process is not customized to identify potential candidates who could be trained easily, as hiring managers are not experienced in sourcing these easily trainable candidates. This causes deficiencies in building up a steady talent pipeline.
11. Creating learning opportunities: Encouraging employees to contribute to technical white papers on AI topics, participating in knowledge sharing across various AI journals, participating in roundtables and working with industry analysts are some of the other avenues to create learning opportunities.

Top skill sets most suited for transitioning to AI roles

Reskilling/upskilling will ensure adequate scaling of enterprise AI and leveraging transferable skills that are relatable to AI. Today, the top transferable skills for an AI career are linear algebra, probability, statistics, ML algorithms, data science, programming, AIOps, text analytics, image analytics and data mining.

In general, mathematics plays an important role in AI, and specifically in ML. Skills in applied mathematics in the areas of linear algebra, probability theory and statistics, multivariate calculus, algorithms and optimization are particularly relevant. As ML works with huge amounts of data, data science competencies help in predictive analytics, data modeling, analytics and other aspects of AI. There are also multiple programming languages to cater to the algorithms, libraries and frameworks in AI that cover text analytics, image analytics, compute deep learning and neural networks.

Fixing the skills gap by focusing within the organization and bringing about internal transformation will take some patience and conscious effort. But this is an investment worth making, as creating a robust talent pool and pipeline will be one of the primary requirements for seizing the opportunities that the next generation of the AI revolution will provide.

Balakrishna DR, popularly known as Bali, is the executive vice president and head of the AI and automation unit at Infosys.