Chaudhry Faisal

Lead Artificial Intelligence and Machine Learning Architect with 12 + years of hand-on experience designing and delivering scalable machine learning, data engineering, and generative AI platforms in production environments. Specialized in LLMs, Retrieval-Augmented Generation (RAG), agentic AI, and multi-agent systems, with hands-on expertise in building autonomous AI workflows using modern frameworks and orchestration tools. Proven track record of architecting end-to-end MLOps and LLMOps platforms using Kubernetes, Kubeflow, and MLflow across AWS, Azure, and GCP, significantly reducing deployment cycles and improving system scalability.

Strong background in NLP, computer vision, and large-scale data systems leveraging Apache Spark, Kafka, and lakehouse architectures. Experienced in leading high-performing engineering teams, driving AI innovation, and implementing enterprise-grade, secure, and compliant AI solutions aligned with Responsible AI standards.

AI/ML Architect at AltaML (2022-07 – Present)

• Architected and deployed a production-grade agentic AI platform using LangGraph, CrewAI, AutoGen, and OpenAI Agents SDK, enabling autonomous multi-agent orchestration, planning, and tool-calling workflows.
• Designed and implemented generative AI systems leveraging LLMs (Llama 3, Mistral, GPT-4o) with advanced RAG pipelines, embeddings, and vector databases (Pinecone, Weaviate, FAISS) integrated with LlamaIndex and LangChain.
• Built and optimized end-to-end MLOps and LLMOps pipelines using Kubernetes, Kubeflow, MLflow, Argo Workflows, Helm, and Terraform across AWS and Azure, enabling scalable CI/CD, model versioning, and automated deployments.
• Led development of real-time AI systems for NLP and computer vision, utilizing streaming architectures (Kafka) and low-latency inference pipelines in production environments.
• Implemented robust model and LLM observability frameworks using Prometheus, Grafana, LangSmith, LangFuse, and Evidently for evaluation, tracing, drift detection, and performance monitoring.
• Designed feature stores and data pipelines supporting ML and GenAI systems, leveraging Spark, Delta Lake, and lakehouse architectures for scalable data access and consistency.
• Mentored and guided engineering teams on AI/ML system design, MLOps, LLMOps, and distributed systems, fostering best practices and high-performance engineering culture.
• Established and enforced Responsible AI and governance frameworks, aligning with ISO 42001, ensuring compliance, explainability, auditability, and secure AI system deployment.

Senior AI/ML Engineer at Symend (2018-04 – 2022-06)

• Led migration from monolithic ML systems to containerized MLOps platforms using Docker, Kubernetes, Helm, Jenkins, and GitHub Actions, improving scalability, reproducibility, and CI/CD automation.
• Designed and deployed deep learning and NLP models using PyTorch and TensorFlow, leveraging architectures such as BERT and early GPT variants with Hugging Face Transformers.
• Built and productionized end-to-end ML pipelines including data ingestion, feature engineering, model training, validation, and deployment using Pandas, Scikit-learn, and Apache Spark.
• Developed scalable data orchestration workflows using Apache Airflow, integrating ETL/ELT pipelines and supporting large-scale batch and streaming data processing.
• Implemented A/B testing, experimentation platforms, and model evaluation pipelines, enabling data-driven optimization of models and product features.
• Introduced model monitoring, logging, and observability solutions using ELK stack, Prometheus, and Grafana, ensuring reliability and proactive anomaly detection.
• Collaborated cross-functionally to translate business requirements into scalable ML architectures, aligning with product, engineering, and data teams.
• Contributed to MLOps best practices, including model lifecycle management, versioning (MLflow), and reproducibility, improving team productivity and system robustness.

Machine Learning Engineer at Osedea (2015-01 – 2018-03)

• Designed and deployed machine learning models for computer vision and time-series forecasting using TensorFlow, Keras, and Scikit-learn in production systems.
• Developed fraud detection and predictive analytics solutions using supervised and unsupervised learning techniques, improving risk assessment for financial clients.
• Built scalable distributed data processing pipelines using Apache Spark, Hadoop, AWS SageMaker, and integrated storage systems (S3, HDFS).
• Performed advanced statistical analysis, feature engineering, and model evaluation, supporting data-driven product and business strategies.
• Developed reusable ML components, pipelines, and feature engineering templates, improving development efficiency and standardization.
• Collaborated with cross-functional teams to integrate ML systems into cloud-based architectures (AWS, GCP), ensuring scalability, reliability, and performance.

Software Engineer at Osperity (2013-08 – 2014-12)

• Developed scalable backend systems and data ingestion pipelines using Python, Java, SQL, and MongoDB, supporting high-throughput applications.
• Designed and optimized ETL/ELT pipelines and data warehousing solutions, enabling structured data access for analytics and reporting.
• Built early-stage predictive analytics and ML prototypes using Scikit-learn and R, contributing to initial ML adoption within the organization.
• Implemented data modeling, reporting systems, and dashboards, enabling business intelligence and decision support systems.
• Worked in Agile/Scrum environments, contributing to scalable system design, microservices foundations, and software engineering best practices.

Master in Science – University of Sargodha