Software engineers have always been considered immune to job losses or irrelevancy perspectives, but the launch of agentic AI is changing that. This advanced generative AI technology is already working alongside experienced software developers to get tasks done faster. It made Salesforce contemplate freezing its hiring for software engineers in 2025. Here’s everything we know about it.
What are AI Agents and Agentic AI?
Agentic AI is a sophisticated system/program capable of autonomously performing tasks on behalf of users or another system. It comprises an LLM that orchestrates the behavior of multiple AI agents that can be effortlessly deployed across various applications. AI agents are designed to handle complex AI applications that traditional AI cannot, making it a crucial part of an organization’s process automation strategy.
Tech giants like Salesforce are incorporating this technology into their work processes because it allows them to make decisions without external support and take appropriate actions. It can also solve complex problems and interact with the external environment beyond the trained data, facilitating faster work completion.
Salesforce’s Decision to Freeze Hiring
In a podcast, Marc Benioff, Salesforce’s CEO, revealed that the company may not hire any software engineers in 2025. This decision comes after noticing the productivity gains from using AI agents in their internal workflow systems. Instead of hiring more software engineers, Salesforce plans to deploy AI agents. These agents will work alongside their existing engineers for greater productivity.
Besides productivity gains, Marc cited the global labor shortage as a crucial reason behind the widescale adoption of AI agents. The CEO believes hiring people in sales and service is challenging, especially in the United States. In such scenarios, AI agents will enable businesses to do more.
Other Tech Giants Embracing this Shift
Salesforce isn’t the only tech giant turning to AI agents instead of human software engineers. In a Joe Rogan podcast, Mark Zuckerberg, Meta’s CEO, revealed that Meta and other companies will work on deploying AI that takes the position of mid-level engineers to write codes in 2025. While he admits that this step will be expensive initially, he thinks it will become cost-efficient over time. According to Mark, there will come a point when the code in apps, including AI-generated ones, will be built by AI engineers instead of their human counterparts.
Jensen Huang, NVIDIA’s CEO, also thinks there’s a huge possibility that in the next 24 months or some other amount of time, most developers won’t be coding. He even recommended that farming, manufacturing, biology, and education may be more plausible career options for the next generation. Microsoft CEO, Satya Nadella, also envisions a future where organizations will use a swarm of AI agents for their work, reducing the need to hire more human engineers.
Reasons Behind Widescale Adoption of Agentic AI
The biggest reason companies embrace agentic AI is its ability to speed up most of the work done by human software engineers. This technology can enhance the output quality, boost the team’s productivity, and get more done in a relatively shorter period.
One standout feature of this technology is its ability to automate tasks and make independent decisions. This autonomy can lower operational costs while improving overall performance. Organizations also prefer agentic AI because it frees up human teams by automating repetitive tasks and streamlining workflows so they can focus more on high-value, strategic work.
Potential Ways in Which Companies May Use Agentic AI
Tech giants can use agentic AI in numerous ways, ranging from autonomous testing to code completion. Instead of spending days finding issues in their code, they can use agentic AI to generate code snippets for faster error identification.
Companies can use AI agents for automated code writing, code suggestions, bug detection, automated testing, automated code review, refactoring, and more. These are a few potential use cases of agentic AI or AI agents as virtual software developers, but organizations can explore endless ways that align with their requirements.
Downsides of Using Agentic AI
While agentic AI has plenty of upsides, organizations cannot let it run without adequate human supervision. Since these systems are free to make independent decisions, it may sometimes lead to unintended consequences.
Concerns around agentic AI potentially misusing user data also exist, calling for stricter guidelines around transparent data collection, storage, and usage practices. If agentic AI makes decisions based on biased data that it was trained on, then organizations can have a challenging time mitigating damages resulting from discriminatory or harmful outcomes.
Impact on Human Software Engineers
While the benefits and capabilities offered by agentic AI may make people believe that human software engineers aren’t required anymore, it isn’t true. This technology isn’t without its challenges, so companies will always need talented software engineers.
The biggest impact on these professionals will be ‘refinement’ not ‘obsolescence.’ Software developers who level up their skills and learn to collaborate with AI agents will thrive in the coming years, while those refusing upskilling will become irrelevant in the hiring landscape.
Way Ahead for Software Engineers
Software engineers can expect AI agents to transform their jobs instead of replacing them entirely. AI agents are here to stay and will keep evolving with time, improving with every upgrade. In such scenarios, software engineers will have to change their work approach to remain relevant. Instead of generating code, they’ll be expected to guide through AI systems, helping the machine generate high-quality output. This change will shift the focus from traditional coding tasks to strategic, human-centered tasks.
