Entry Level AI Software Engineer

About the Role

At IBM Software, we transform client challenges into solutions by building AI-powered, cloud-native products. We foster a culture of curiosity, collaboration, innovation, growth, and continuous learning.

Your Role And Responsibilities

As an Entry Level AI Software Engineer at IBM Z, you will help bring AI to the systems that support global commerce, banking, and payments. You will work alongside experienced engineers to turn advances in large language models (LLMs) into useful software. Depending on your team, you might:

• Build an agent to diagnose system issues.
• Connect AI to enterprise data and tools.
• Create simpler ways for developers to work with complex systems.
• Bring your ideas, experiment, and help take a feature from prototype through testing and refinement.

What You’ll Do

• Design, build, and test AI-powered features that address real developer and client needs, with guidance from experienced engineers.
• Integrate LLMs, APIs, and enterprise data into applications, exploring agent workflows, tool calling, and retrieval.
• Write clear specifications and effective prompts, use AI development tools thoughtfully, and review and test the code they produce.
• Evaluate AI behavior with realistic test cases, measure quality and performance, and improve how applications handle incorrect outputs and failures.
• Debug across model interactions, application code, APIs, and underlying systems to understand root causes and build reliable solutions.
• Collaborate through Git, code reviews, and technical discussions; document your work and iterate on feedback.

Requirements & Tech Stack

• Strong software engineering fundamentals, including data structures, algorithms, software design, and testing.
• Proficiency in Python and experience building applications or services that integrate APIs.
• Hands-on experience building an application or working prototype with LLMs. You should be able to explain what you built, how you integrated the model, and how you tested its behavior. Experience may come from personal, academic, open-source, or internship projects.
• A systematic debugging mindset: ability to investigate failures, test assumptions, and reason about interactions between application code, data, models, and underlying systems.
• Ability to translate a problem into clear requirements, prompts, and testable acceptance criteria, and critically evaluate AI-generated code and responses.
• Familiarity with Git, clear technical communication, and the ability to learn quickly and collaborate on a team.

Preferred Qualifications

• Experience in one or more of the following areas is a plus.
• AI agents and tool or function calling; retrieval-augmented generation (RAG), embeddings, vector databases, or Model Context Protocol (MCP).
• Independent AI projects, open-source contributions, or experiments beyond assigned coursework.
• Additional programming experience in JavaScript/TypeScript or Go, and familiarity with distributed systems or microservices.
• Operating system concepts such as concurrency, multithreading, and memory management; relational databases and SQL.
• Cloud and container technologies such as Docker, Kubernetes, or OpenShift; observability, search, or real-time data processing tools such as OpenSearch or Kafka.
• AI evaluation, anomaly detection, or experience measuring and improving model accuracy, latency, and reliability.

We are especially interested in what you have built and learned. Where available, include a project link, repository, demo, or a brief description of your contribution and the technical challenges you worked through.

Education

• Bachelor's Degree