I'm an AI developer with 4+ years of experience building production NLP and retrieval systems. I hold an M.Sc. in Artificial Intelligence & Robotics and focus on making large language models genuinely useful — grounded, retrieval-augmented, and connected to real tools.
My recent work includes semantic search platforms for large-scale religious text corpora, RAG pipelines with semantic chunking and LLM propositioning, and AI agents that operate geospatial services. I've fine-tuned BERT and Whisper models for Persian, built NER and OCR post-correction systems, and shipped them all as scalable APIs.
I also published research on measuring semantic similarity of Persian sentences using ParsBERT (2024).
# ali.py class AliZareshahi: role = "AI Developer" focus = ["RAG", "Agents", "Persian NLP"] degree = "M.Sc. AI & Robotics" years = 4 def build(self, idea): return production(idea)
Building AI-powered products with LLM-based pipelines and retrieval systems.
Developing agentic AI features and integrating LLMs into production services.
Built RAG chatbot infrastructure for business knowledge bases.
Semantic search and NLP services over large-scale religious text corpora; fine-tuned language models for Persian and Arabic content.
NLP pipelines including NER and OCR post-correction for digitized archives.
Research internship on Persian text processing and semantic similarity.
Built web front-ends; foundation in JavaScript and modern web development.
Semantic similarity lab & AI-powered search over Hadith and religious texts — embedding-based retrieval at corpus scale.
RAG platform with semantic chunking and LLM propositioning for accurate, grounded answers over custom knowledge.
Web-based geospatial data management platform with map visualization and AI agents driving GeoServer WPS services.
Research on transformer-based semantic similarity for Persian, applied to large-scale retrieval of religious texts.
Whether you have a project involving RAG, agents, or NLP — or just want to say hi — my inbox is always open.