Principal Engineer (AI Systems)

We are looking for a hands-on Principal Engineer with expertise in AI Systems who is passionate about developing scalable, production-ready Generative AI solutions. This role requires active coding, experimentation with large language models (LLMs), and turning innovative concepts into practical applications. You will focus on Retrieval-Augmented Generation (RAG), LLMOps pipelines, and multi-agent orchestration frameworks, tackling complex technical challenges daily. As a purely technical individual contributor, you will lead by example through writing code, mentoring peers via reviews, and managing the full lifecycle of technical delivery.

Key Responsibilities
Design and develop RAG systems that leverage embeddings, hybrid search techniques, and evaluation pipelines. Build and maintain multi-agent orchestration frameworks such as LangGraph, AutoGen, CrewAI, or custom-built solutions. Implement and oversee LLMOps pipelines for prompt versioning, cost tracking, and performance evaluation. Integrate AI workflows seamlessly with backend services and data layers to ensure scalability in production environments. Experiment with LLMs to improve retrieval, summarization, and personalization use cases. Contribute directly to codebases, participate in architecture reviews, and drive performance optimizations. Collaborate closely with data and platform engineering teams to deploy and optimize Generative AI solutions.

Required Qualifications
A minimum of five years’ experience in backend or machine learning engineering with strong proficiency in Python programming. Proven experience delivering RAG systems involving vector databases, embeddings, and data chunking. Familiarity with orchestration frameworks such as LangGraph, LangChain, AutoGen, or similar tools. Solid understanding of LLM behavior, evaluation methodologies, and fine-tuning workflows. Experience working with APIs, microservices, and cloud-native development, preferably on AWS.

Preferred Qualifications
Experience handling unstructured data formats including PDFs, tables, and images. Knowledge of distributed systems concepts such as asynchronous processing, message queues, and caching mechanisms. Exposure to LLM evaluation techniques or reinforcement learning from AI feedback (RLAIF). Understanding of data versioning practices and retrieval metrics.

Soft Skills
A builder mindset with a passion for writing, debugging, and refining production-quality code. Collaborative and humble, open to feedback and continuous learning. Strong communication skills with the ability to clearly articulate design decisions. Ability to influence team outcomes through technical contributions rather than formal authority.

Emumba is committed to delivering innovative solutions and exceptional services tailored to the diverse needs of our clients. We prioritize quality and customer satisfaction, striving to exceed expectations and drive success in every project.

Department: Backend
Employment Type: Full Time
Location: Islamabad, Pakistan
Workplace Type: Fully remote