AI Product Engineer · Founder · Open-Source Builder
Turning AI into products people actually use.
I'm an AI product engineer and founder who takes AI from idea to production. I work where models meet real users: shaping the product, designing the system, and shipping software that people rely on every day.
My focus is simple: AI should be judged by the problems it solves, not by the demo it makes.
- AI & LLM products: turning large language models into reliable, shipped products with clear user value
- Agentic workflows & developer tooling: agents, skills, rules and context systems that make engineers and teams faster
- Strategy to production: from product discovery and roadmap to architecture, launch and iteration
- E-commerce & marketplaces: search, discovery and recommendation for buyers and sellers
- Tech leadership & digital transformation: helping teams and businesses adopt AI in ways that last
- Open source: building in public and sharing the tools behind the products
Production-grade building blocks for AI-powered commerce, built in the open.
| Project | What it does | |
|---|---|---|
| woocommerce-mcp |
Model Context Protocol server for WooCommerce: multi-store, previewed safe writes, refunds, bulk pricing and sales analytics. | Docs |
| llm-search-eval |
Measures e-commerce search quality with IR metrics and LLM judge panels, with significance tests, cost budgets and a CI quality gate. | Docs |
| persian-rag |
Persian-first RAG toolkit for Python: Farsi normalization, PDF and Word loading, hybrid retrieval, reranking and evaluation. | Docs |
| commerce-agent-skills |
Skills and rules that help AI coding agents build e-commerce and marketplace products the right way. | Docs |
- persian-rag-bench: A Persian Retrieval Benchmark for Spelling-Variant Robustness on Figshare, 2026
- persian-rag: A Persian-First Retrieval-Augmented Generation Toolkit — Normalization, Hybrid Retrieval, and Retrieval Evaluation on Zenodo, 2026
- Technical Report: WooCommerce Model Context Protocol Server - A Multi-Store, Previewed Safe-Write Architecture for Autonomous AI Agents in E-Commerce on Zenodo, 2026
- commerce-agent-skills: Architectural Specification, Domain Constraints, and Evaluation Framework for E-Commerce AI Coding Agents on Zenodo, 2026
- A week of reader feedback: what changed in my four AI-for-commerce tools on Dev.to
- Safe write tools for agents: preview, confirm, audit, and stale previews in a WooCommerce MCP server on Dev.to
- Why RAG breaks on Persian text, and what I built to fix it on Dev.to
- Is your new search actually better? Label gaps, LLM judge panels, and paired statistics on Dev.to
- Treat agent skills like code: evals, a strict validator, and an npx installer on Dev.to
- What reader feedback taught me about building production-safe AI tooling for commerce on LinkedIn
- Preview, confirm, audit: designing safe write tools for AI agents on a live store on LinkedIn
- Is your new search actually better? Five questions I ask before shipping a ranking change on LinkedIn
- Why RAG quietly fails on Persian text, and the design decisions that fixed it on LinkedIn
- The knowledge your senior engineers carry, packaged for coding agents on LinkedIn
- Four open-source tools for AI in e-commerce, and what building them taught me on Dev.to
Linked items point to the open-source project that uses them.
AI & LLMs: OpenAI · Anthropic · Gemini · Llama · Mistral · Hugging Face Transformers · Ollama · vLLM
Agents & orchestration: LangChain · LangGraph · LlamaIndex · CrewAI · AutoGen · OpenAI Agents SDK · MCP (Model Context Protocol)
RAG & retrieval: Embeddings · Pinecone · Weaviate · Qdrant · Chroma · pgvector · Rerankers · Elasticsearch · Algolia
ML & evaluation: PyTorch · scikit-learn · Fine-tuning (LoRA) · LLM evals · Prompt & context engineering · Guardrails
AI developer tools: AI coding agents · Agent skills & rules · Prompt tooling · Observability (LangSmith, Langfuse)
Languages: Python · TypeScript · JavaScript · PHP · SQL
Backend & APIs: Node.js · FastAPI · Laravel · REST · GraphQL · WebSockets
Frontend: React · Next.js · Tailwind CSS
Data: PostgreSQL · MySQL · Redis · MongoDB
E-commerce: WooCommerce · WordPress · Payment integrations · Marketplace search & recommendations
Cloud & delivery: Docker · Kubernetes · GitHub Actions · CI/CD · AWS · Linux · Nginx
- LinkedIn: in/sadrmanesh
- X: @nasersadrmanesh · Instagram: @nasersadrmanesh · Threads: @nasersadrmanesh
- Writing: Dev.to
- Hugging Face: sadrmanesh
- Mastodon: @sadrmanesh@mastodon.social
- Research: ORCID · Google Scholar
ساختن محصول با هوش مصنوعی، برای آدم های واقعی.



