Senior Software Engineer (AI/ML) - Part-time | Remote

About Us CloudPSO is a Information Technology Outsourcing (ITO) company that assists in the acquisition of qualified staff to address complex digital problems in order to increase efficiency, reduce costs, and maintain compliance. CloudPSO was founded in 2017 with an aim to provide businesses with a competent and skilled workforce at any given point in time and from any geographic region. We are a US-based company with headquarters in Dallas (Texas) and a center of excellence in Pakistan. We have over 200 facility seats with an additional Work-From-Home facility. CloudPSO has skillful in-house software development teams with state-of-the-art tools, the latest VOIP technology platform, and secure infrastructure. Our core values consist of client satisfaction, commitment, quality, and transparency. We, at CloudPSO, hunt, analyze, recruit, train, and retain top-notch talent for you to help achieve your business goals. Optimizing mission-critical and day-to-day enterprise IT operations, CloudPSO enables businesses to transform, innovate and scale. Job Description This is a remote position. Job Location: Pakistan (Remote) Employment type: Part-time Job timings: US time zone We are looking for a Senior Software Engineer (AI/ML) who can design, build, and scale intelligent systems — from machine learning pipelines to advanced AI agents capable of reasoning, planning, and automation. You will work on building production-grade AI models, integrating LLMs and tools, and designing agent architectures that interact with APIs, databases, and workflows. The role blends applied ML expertise with strong backend engineering and product-focused problem-solving. Responsibilities: Design and build autonomous or semi-autonomous AI agents that can plan, reason, and interact with tools, APIs, or external systems Implement agentic frameworks (e.g., LangChain, LlamaIndex, CrewAI, or custom orchestration systems) leverage existing industry capabilities to deliver virtual assistant capabilities on top of xquic content including voice interactions Optimize reasoning and retrieval pipelines using embeddings, vector databases, and prompt engineering Develop, train, and fine-tune ML models using frameworks like PyTorch, TensorFlow, or scikit-learn Work on data preprocessing, feature engineering, and model evaluation for NLP, computer vision, or predictive tasks Build ML pipelines for training, deployment, and monitoring in production environments Collaborate with engineering teams to integrate AI components into backend systems and APIs Ensure scalable, maintainable codebases with CI/CD, observability, and cloud-native design (AWS/GCP/Azure) Contribute to technical architecture and design reviews for AI-driven features and platforms Stay current with the latest in LLMs, agent frameworks, and model architectures Prototype and evaluate new approaches for reasoning, tool use, and adaptive behavior in agents Share learnings and mentor peers in ML and AI development best practices. Requirements Bachelors Degree in Computer Science or related fields 5 years of experience Strong programming skills in Python (mandatory); proficiency with PyTorch, TensorFlow, or transformers-based models Experience in building or integrating AI agents (LangChain, LlamaIndex, CrewAI, custom frameworks) Strong grasp of ML model lifecycle — data processing, model training, evaluation, deployment, and monitoring Experience with cloud platforms (AWS/GCP/Azure) and containerization (Docker, Kubernetes) Familiarity with API integrations, microservices, and asynchronous systems Strong understanding of vector databases (e.g., Pinecone, Weaviate, FAISS, Chroma) and retrieval architectures Solid software engineering fundamentals — testing, version control, and system design Experience with LLM fine-tuning, prompt optimization, or RAG (Retrieval-Augmented Generation) systems Familiarity with multi-agent systems and coordination mechanisms Knowledge of MLOps tools (MLflow, Kubeflow, Sagemaker, Vertex AI) Experience in AI system observability, evaluation metrics, and continuous learning loops Exposure to reinforcement learning or self-improving agent architectures