Shahriar Shariati
AI Engineer & CS Researcher
Building intelligent systems that bridge research and production. I specialize in LLMs, NLP, and agentic AI systems — turning cutting-edge research into production-ready solutions.
Bridging Research & Production
MSc student in Computer Science (Data Mining) at Shahid Beheshti University. I focus on making AI systems that actually work — from rigorous research benchmarks to production-ready agentic workflows. My work spans NLP, LLM evaluation, and building multi-agent systems that solve real-world problems.
NLP & Large Language Models
Deep expertise in transformer architectures, fine-tuning, and prompt engineering
LLM Evaluation & Benchmarking
Building comprehensive evaluation frameworks for language models
Agentic AI Systems
Designing and implementing multi-agent workflows for complex tasks
Persian Language AI
Contributing to Persian NLP research and model evaluation
MSc Computer Science
Data Mining Specialization
Shahid Beheshti University, Tehran
BSc Electrical Engineering
Computer Engineering
Ferdowsi University of Mashhad
Work & Research
Building production AI systems and contributing to NLP research
AI Engineer
Rudys.AI
Building LLM-powered agentic workflows and multi-agent systems. Designing and implementing AI solutions that automate complex business processes with intelligent orchestration.
NLP & LLM Researcher
Shahid Beheshti University
Leading research on Persian language model evaluation. Developed MELAC benchmark covering 19 datasets and 41 models. Published at ACL 2025.
Senior Python Developer
IFSGuide
LangChain development, prompt engineering, and Kubernetes deployment for AI-powered enterprise applications.
Senior Back-End Developer
OrgMeter
Built CRM systems with PostgreSQL and microservices architecture. 3+ years of production backend experience with scalable systems.
Publications
Contributing to Persian language AI research and LLM evaluation
Persian in a Court: Benchmarking Large Vision-Language Models
Comprehensive evaluation of Large Language Models and Vision-Language Models on Persian multimodal tasks. Establishing baselines for Persian VLM capabilities.
MELAC: Massive Evaluation of LLMs in Persian
Comprehensive evaluation framework with 19 datasets for Persian LLM evaluation. Benchmarked 41 different language models to establish Persian AI capabilities.
Technical Expertise
Technologies and domains I work with
AI & ML
Frameworks
Backend
DevOps
Let's Build Something Together
I'm passionate about turning AI research into production systems that create real value. Whether you have a project in mind, research collaboration opportunity, or just want to chat about AI — I'd love to hear from you.
Send a Message
Fill out the form and I'll get back to you as soon as possible.