SoftServe, a global IT consulting and digital services provider, is integrating AI agents into its teams, which can reduce development time by up to 70%. AI-based services currently represent the fastest-growing segment within the company.
In just over two years of activity, the GenAI Lab team at SoftServe has developed more than 200 AI-based solutions for over 100 clients. AI-driven services are the fastest-growing segment within the company, recording an annual growth rate of 85%. The highest demand is currently for agentic systems.
„We are among the few companies in the world that are already implementing full development automation through agentic engineering in client projects — from requirements gathering to testing. Our strategy has three directions: we create GenAI solutions for clients, use agents to accelerate our own development, and integrate AI into corporate processes. Thanks to our partnerships with NVIDIA, AWS, Google Cloud, and Microsoft Azure, we have access to the latest technologies and can deliver measurable results where others are still in the experimentation phase.”, explains Serge Haziiev, Advanced Technologies CTO în cadrul SoftServe.
Last year, the industry’s attention was focused on chatbots and retrieval-augmented generation (RAG) systems. Now, it has shifted toward agentic AI and is gradually advancing toward Physical AI — systems capable of perceiving and interacting with the real world.
AI Agents in Action: From Prototypes to Large-Scale Implementation for Clients
The company’s developers have access to a wide range of AI tools (ChatGPT Enterprise, Cursor, Midjourney, and others), as well as to internal agents. SoftServe’s portfolio already includes more than a dozen proprietary AI agents that automate project planning, analysis, testing, and engineering tasks. The company is also developing its own Agentic Intelligence Open Platform, designed to simplify the development process and enable more efficient implementation of AI systems.
SoftServe is already deploying AI agents within client projects. Depending on the project, their use can reduce development time and costs by 30% to 70%.
„We are not aiming for people to use AI merely as another tool or application. Instead, we seek effective collaboration between developers and their virtual colleagues — enabling outcomes that were previously unattainable.”, adds Serge Haziiev.
GenAI Lab develops agents that automate and accelerate the software development process, as well as specialized agents for specific tasks. These include platforms for the pharmaceutical industry, hyper-personalized shopping assistants, and production support assistants. SoftServe deploys its solutions in any environment — from the cloud to clients’ on-premises servers — using technologies from NVIDIA, AWS, Google Cloud, and Microsoft Azure.
SoftServe has introduced the role of Intelligence Engineer.
Over 150 SoftServe specialists are currently working on AI implementation projects, forming cross-functional teams focused on LLMs, agentic engineering, and AI-assisted development. Despite the growing demand, the market faces a severe shortage of highly skilled talent, particularly in Data Science. According to a World Economic Forum report, AI specialist roles are expected to be among the fastest-growing by 2030.
In this context, SoftServe has already introduced a role dedicated to this new era — the Intelligence Engineer. This specialist acts as an interface between human expertise and agent-generated results: orchestrating AI agents, validating their performance, and ensuring proper integration. The GenAI Lab team already includes 40 such professionals, and the company plans to double this number within the next year.
The company is also seeking engineers with AI expertise in Romania.
The Skills of the Future
According to Serge Haziiev, the three key competencies that will remain vital in the market are expertise in various industries and business processes, architectural design of IT systems, and AI agent orchestration.
The company is actively preparing its teams for these changes: over 50% of employees, most of them from delivery departments, have already completed AI training courses and gained the necessary skills to work with these technologies.
In SoftServe Romania, AI-focused training programs have also gained momentum, with learning hours dedicated to AI increasing twelvefold in Q3 2025 compared to Q1.
Depending on the skill level, SoftServe offers multiple learning paths — from basic courses to advanced technical specializations. Employees can also obtain international certifications from NVIDIA, AWS, and Google Cloud Platform, or participate in AI hackathons and internal technical workshops.
