AI Solutions Engineer
Pand.ai · Singapore
Apply & track with Apply EdgeAI Solutions EngineerSingapore | Full-time | 3-5 Years ExperienceABOUT PAND.AIPand.ai builds enterprise AI systems for some of the largest organisations in Asia. Our engineers work on real production problems where AI must integrate with existing enterprise systems, operate within security and regulatory constraints, and deliver reliable outcomes at scale.Because we work closely with our customers, our engineers don't just receive specifications. They learn to understand business problems, uncover technical constraints, and design and build complete AI systems.THE ROLEAs an AI Solutions Engineer, you will own AI systems end to end, from the customer conversation that defines the problem through to the code running in production.This is an engineering role, not a pre-sales one. You will write the code that ships. But you will also be in the room with the customer, because the hardest part of enterprise AI is usually working out what should be built rather than building it.You do not need to have worked on AI systems before. We are looking first for strong software engineering fundamentals, sound judgement under ambiguity, and the ability to learn quickly.But this is an intellectually demanding role with a steep learning curve. Many of the problems we work on do not have established playbooks. You will regularly encounter technologies and problems you have never seen before.Because we are a lean team, you will get unusually broad technical exposure and real ownership of the systems you build. You will learn through code review, design discussions and hands-on exposure to difficult real-world problems, and you will see the commercial side of the work, including why enterprise constraints shape what gets built.In return, we expect you to take ownership of your learning and develop quickly.WHAT YOU'LL WORK ONDesign and build AI-powered applications, backend services and APIs using PythonBuild RAG, semantic search, agentic and LLM-powered systemsIntegrate AI with enterprise data, APIs, tools and business workflowsBuild product interfaces using React and TypeScriptDeploy and operate applications using Docker and cloud infrastructureDesign, debug and improve systems for reliability, security and performanceWork directly with enterprise customers to understand technical and business constraints and translate ambiguous requirements into engineering solutionsWHAT WE'RE LOOKING FORBachelor's degree in Computer Science, Software Engineering, or a related engineering or quantitative field3-5 years of professional software engineering experienceVery strong Python. This is the core requirement, and we will test itTrack record of building production software, backend services or APIs that other people depended onGood understanding of system design, databases, APIs and the trade-offs between themStrong analytical, problem-solving and debugging abilityAble to learn unfamiliar technologies quickly and tackle problems where the answer is not immediately obviousAble to hold a technical conversation with a customer, including when the answer is noTakes ownership of outcomes rather than tasksWe care less about which frameworks are on your CV and more about the strength of your fundamentals and whether you can take an unfamiliar problem, understand it quickly, reason about it clearly and figure out how to solve it.USEFUL, BUT NOT REQUIREDExperience with LLM applications, RAG, agentic AI, React/TypeScript, Kubernetes, cloud infrastructure or machine learning. Prior work in financial services, insurance or legal technology.WHO WILL THRIVE HEREIf you want to master a fixed technology stack and then rely on what you already know, this is probably not the role for you.There are easier engineering jobs. This one is for someone who wants real ownership, harder problems than they currently have, and to become a much stronger engineer.OUR PROCESSShortlisted candidates will complete a Python assessment before the technical interview. We use it to see how you reason about code, not to check whether you have memorised syntax.