In my work architecting AI transitions for global enterprises, I have identified a recurring systemic failure, a collision between hyper-accelerated output and stagnant governance. IT & Software Services (IT companies) are currently trapped in a talent paradox where they have an oversupply of AI-augmented ‘coding speed,’ yet a critical shortage of ‘architectural safety.’ They have effectively commoditized the ‘build,’ but in doing so, they have made accountability the most expensive resource in the room (developers/engineers → AI-augmented coding speed, project managers → AI-augmented delivery efficiency, solution architects → AI-augmented design quality, technical architects → AI-augmented system reliability, business analysts → AI-augmented requirements clarity, testers → AI-augmented defect detection rate, etc.).
Currently, the workforce in IT companies is imbalanced, with both a surplus and a shortage of engineers at all levels. Because most primary tasks are being automated, yet these engineers are starving for ‘Performance Architects’ who can govern those very automations. Therefore, managers must treat our oversupplied resources as a talent reservoir, aggressively re-skilling them to manage the ‘hidden costs’ of AI, turning our legacy developers into the AI-governance experts the market is desperately lacking.
An oversupply of labor exists as engineers lack specialized skills
The labor market is oversupplied because engineers mainly possess general skills in the field of IT in general and software outsourcing in particular represents one of the most pressing challenges in workforce planning. In the past five years, since the emergence of generative AI, universities and training institutions have produced a large number of graduates with broad IT competencies, such as generic software development, basic database management, cloud computing concepts, cybersecurity fundamentals or basic networking. While these skills are foundational, they often lack the specialization required to address the complex, evolving needs of modern IT corporations. As a result, many professionals enter the workforce with overlapping capabilities, leading to inefficiencies in hiring and deployment for mass projects or outsourcing for complex business domains.
In IT companies (India, Philippines, Vietnam, Poland and Ukraine), this oversupply creates a paradox: Organizations have access to a large talent pool, yet still struggle to fill critical roles or workforce assurance (WA) division is unable to provide sufficient manpower when dealing with large-scale and technologically complex projects from clients. For example, a business unit (BU) has a big project of client in banking and finance services, the BU may need hundreds of applicants for a software developer position, but very few candidates with expertise in smart-based secure payment systems or international regulatory compliance frameworks. The mismatch leads to underemployment for the current BU’s workforce, and even some individuals are assigned to roles that fail to utilize their full abilities, or worse, remain unemployed despite possessing professional qualifications.
The consequences of short-term mismatch skills in the IT field extend beyond career path development for individuals. The WA division in IT companies is facing a large idle labor pool, bloated recruitment pipelines, higher turnover rates and wasted resources on retraining as well. Moreover, the oversupply of generalists dilutes the competitive edge of the industry in the IT outsourcing market, as innovation increasingly depends on niche expertise. Without long-term corrective approaches, such as rebalancing educational curricula for internships, enriching skilling for oversupply resources, filling gaps in programs for potential engineers, or incentivizing specialization, the growth rate of the sector risks stagnation.
To address this issue systematically, comprehensively and without overlap, coordinated actions are needed among human resources, workforce assurance, learning & development (L&D) and policymakers to ensure that the company has professionals with both breadth and depth of skills.
Short-term shortage of nickel skills and their root causes
The shortage of specialized IT skills has far-reaching implications for the growth and scale extension. It also slows productivity initiatives, increases reliance on external resource vendors and exposes organizations to regulatory and operational risks. Moreover, competition for scarce internal talents between BUs and the labor market drives up salaries, creating inequities and making it difficult for workforce allocation and the cost of enriching skills programs.
In most IT companies, one of the most persistent challenges in talent development is the mismatch between education and industry demand, not only for AI-augmented IT services but also for relevant skills in the AI-driven business domain. Academic curricula often lag behind technological advancements, especially in the AI ecosystem, leaving graduates ill-prepared for the realities of modern banking. For instance, while universities may still emphasize legacy programming languages or outdated IT business models, the industry increasingly requires expertise in AI-augmented engineering, agentic AI, cloud computing, digital transformation and advanced analytics.
Upon being hired by the company, this lag creates a widening skills gap. Interns enter the workforce with knowledge that is already obsolete, forcing companies to invest in retraining. Meanwhile, emerging technologies such as quantum computing, semiconductors, self-driving cars, flying cars, or decentralized finance remain absent from most curricula. The serious mismatch also discourages innovation, as junior engineers are not exposed to cutting-edge tools or real-world applications during their studies. The consequences are significant. BUs are facing delays in implementing new technologies for clients, as they must wait for employees to acquire relevant skills.
Resource allocation inefficiency and issues of turnover rates
In the competitive labor market today, recruitment is particularly costly. Instead of accelerating employee professional development through various methods available on the market, IT companies often compete for the same pool of specialized talent, which inadvertently drives up salaries and recruitment fees compared to the industry average. However, many hires leave within a few months or years, forcing organizations to repeat the cycle.
Enriching (refreshing or retraining) is equally model and same expensive, as employees must be upskilled or reskilled to meet evolving demands, while clients have requests for upgrading new technologies to their business. However, without a clear alignment between professional training programs and prioritized resource needs for each job level, retraining or enriching skilling efforts are often wasteful and fail to deliver measurable returns.
Resource allocation inefficiency slows innovation, drains financial reserves and undermines clients’ satisfaction, leading to BUs failing to meet revenue targets, as they lack the budgets to sustain repeated review benefits and consider increasing job rank as well as retraining cycles.
Any issues related to workforce size lead to issues related to operating costs, growth rates and the micro and macro management policies of the business. Therefore, resource allocation inefficiency is a hidden but critical challenge in talent development. IT companies spend heavily on recruitment, retraining and retention, yet much of this investment fails to produce sustainable results. The root cause lies in misaligned talent pipelines, where resources are directed toward filling immediate gaps rather than building long-term capabilities.
Enrich skilling strategy and the professional training ecosystem
To solve this challenge, BUs must invest in targeted training programs, AI tools, short-term specialization courses, professional courses on online platforms such as Coursera, edX, edXOps, Datacamp, etc., partnerships with technical universities, and global LABs. Additionally, the head office and overseas branches also play a role by funding specialized training programs and creating certification standards. Without such interventions and reforms, the mismatch between education and industry demand and the shortage of specialized skills will remain a critical barrier to achieving the 2030 roadmap.
As results, the bottleneck of talent development and workforce assurance in IT companies isn’t technical; it’s strategic and organizational. We see the same story in most IT service providers, including the outsourcing sector. The AI team creates a breakthrough, but the project stalls because the people responsible for performance and the people responsible for risk operate in silos. To bridge this gap, the WA division must adopt agile approaches that evolve in enriching skilling tandem with industry needs by treating continuous training, AI performance and governance risk as a single measurable system, ensuring that every component of AI‑generated deliverables is as fiscally sound and legally compliant as it is productive.
The training ecosystem for IT talent is highly fragmented, with universities, bootcamps, online platforms and corporate learning programs operating in silos. Each institution designs curricula independently, often without alignment to specific objective priorities (scientific research team or provider of high-quality human resources) or industry-wide needs (AI, Data, Cloud, Robotic and Automotive engineering, etc.). This lack of coordination creates inefficiencies and weakens the overall readiness of the workforce in the era of an AI-driven economy.
For example, universities tend to emphasize theoretical knowledge, producing graduates with strong academic foundations but limited practical exposure to emerging technologies. Bootcamps, on the other hand, focus on rapid skill acquisition, often prioritizing coding or data analysis without integrating broader industry context. Corporate training programs are tailored to immediate organizational needs, but they rarely scale beyond the company’s boundaries. The result is a patchwork of training initiatives and the lack of synchronization that fail to produce a cohesive talent pipeline that adapt complexity technology stack in mass projects of BUs or high-level requests from clients.
Need consistency principles and futuristic approaches
As a technology expert and IT workforce structure consultant for various technology companies worldwide over the years, I have witnessed these fragmentation and overlap in the workforce development system has led to unexpected results.
First, this is because the person in charge has not categorized the courses according to professional level, specialized skills, business model, project specifics, system structure and client culture. Instead of organizing training courses or providing skill-enhancing courses at the general and basic skills levels (offline, online, hybrid learning), the focus is on specialized and project-based courses, which fall under the training phase for transitioning into the production unit and are the responsibility of their senior engineers. They lead to wasted effort due to excessive, misdirected and mistargeted training. As a result, the training program produces interns whose skills do not meet BU’s requirements.
Second, most IT companies have career path diagrams for each job. By identifying core skills and advanced skills in these diagrams, we can build the most suitable learning path for each job group. The problem is that not all employees in a job group have the same needs for knowledge, skills or experience enhancement. Some may already possess these skills, while others may need to develop themselves to meet unit requirements or serve specific projects. Therefore, allowing employees to choose their own skills is essential. Forcing employees to develop mandatory skills according to a specific learning path will confuse them, forcing them to struggle to navigate between many different paths without clear guidance on which skills are most valuable at any given stage. Therefore, personalizing the learning path is a standard that must be considered for employees in production units. Conversely, it would waste a lot of time for learners, incur costs for instructors (mentors, trainers), cover operating costs (training officers), infrastructure usage costs and investments in online platforms. As a result, the training program produces engineers who have skills that do not meet client expectations or BU’s requirements.
Third, for IT service companies, organizing professional training courses on technology, application functionality, industry standards, etc., at various levels for staffs whose engaged in projects is almost mandatory when BUs receive projects from clients. The development of training curriculum, conducting the training and evaluation of results in this format is usually done by senior engineers within the BU (those with expertise matching the project’s requirements), external trainers (those with expertise matching the project) or experts from the client. Furthermore, the absence of standardized curricula makes it difficult to benchmark skills across the industry. Solving this issue requires collaboration between academia, industry and managers to create integrated frameworks that align training programs with priorities for workforce development strategy. Only then can your company build a workforce capable of meeting the challenges of 2030. Otherwise, the consequence is wasted costs and trainees acquiring insufficient knowledge and skills to complete the assigned project tasks.
Fourth, the human resource development process is a continuous cycle encompassing contextual research, needs assessment, model design, strategic planning, resource recruitment, training implementation and results evaluation to align employee skills with business unit needs and organizational goals. This process fosters synergistic development of individuals and businesses through activities such as skill building, career orientation and creating appropriate learning pathways to avoid waste, increase productivity, retain employees and prepare for future needs. To accomplish this mission, the relevant departments need a consistent strategy, reliable information sharing, avoiding overlapping approaches to BUs, refraining from abuse of power and taking full responsibility not only for providing basic vocational training programs for beginners, advanced training for mid-level engineers and professional-level training, but also for focusing on developing programs to address skill gaps to meet ever-changing demands. If these principles are not applied flexibly, the results of human resource development will not meet the organization’s expected goals.
Ultimately, if managers don’t address these issues at their root, the company’s competitiveness in the IT services market is likely to be weakened, as other technology companies may have better coordinated training ecosystems, putting them in a better position to meet industry needs and customer expectations.
Workforce development strategy for IT excellence
In conclusion, IT companies need to build a future-ready workforce that is specialized and aligned with both corporate goals and industry trends. This strategy bridges the gap between education and practice, ensuring that skills evolve in tandem with emerging technologies such as AI, cloud computing, cybersecurity and data governance.
This article is published as part of the Foundry Expert Contributor Network.
Want to join?