AI, Agents & MLOps
- Distributed training and serving
- LLM systems and inference
- Agent orchestration and tool use
- RAG and vector retrieval
- Experiment tracking and model lifecycle automation
About Me
I am an AI and distributed systems engineer focused on trustworthy machine learning, federated and decentralized training, and production-grade MLOps and LLMOps platforms. My work sits where research ideas meet the systems required to run them reliably.
I build across the full stack of modern AI delivery: distributed training and serving, agentic workflows, retrieval systems, streaming data platforms, and the observability, automation, and platform layers that keep models dependable after launch.
Alongside engineering, I teach, review, mentor, and write about AI engineering and data platforms. This site is where I share practical notes from building systems that need to be secure, scalable, and maintainable in the real world.
Shiraz University
Distributed systems, federated learning, privacy-aware machine learning, and secure decentralized training.
Shahid Bahonar University
Artificial intelligence, data mining, and applied machine learning.
Lead engineering for production AI products built around multi-agent workflows, retrieval-augmented reasoning, speech interfaces, and privacy-aware system design.
Architect end-to-end machine learning operations platforms that automate training, evaluation, deployment, and monitoring for research and product teams.
Research and build distributed learning systems spanning federated, clustered, and decentralized training with privacy-preserving protocols and scalable orchestration.
Design secure collaborative learning approaches that combine encryption, privacy controls, and robust evaluation against common leakage and inference risks.
Develop lightweight coordination methods for non-IID federated settings, including similarity-driven client grouping and neighbor selection without exposing raw model values.
Build distributed training and serving workflows, explore scalable LLM deployment patterns, and support research infrastructure for high-performance computing groups.
Deliver large-scale streaming systems, access-control services, and cloud-native microservices for real-time operational data and IoT workloads.
Create automated ingestion and processing pipelines for analytics, geospatial monitoring, and high-volume behavioral data products.
Teaching assistant and seminar presenter on cloud computing, distributed systems, federated learning, secure machine learning, and real-time data processing.
Reviewer for workshops and conferences focused on trustworthy AI, AI safety, and socially responsible deployment.
Contributor to open-source tooling and author of technical writing on AI engineering, data platforms, and production lessons learned.