Personal details
Muhammad J.
About
Iâm an AI Systems Architect and Engineering Lead with 10+ years of experience designing and delivering scalable, production-grade software systems. My background spans full-stack and backend engineering, distributed systems, microservices, cloud infrastructure, data platforms, and technical leadership, with my recent work focused heavily on AI-powered products, RAG architectures, agentic workflows, semantic retrieval, and production LLM integrations.
I combine hands-on engineering with end-to-end architectural ownership. In recent roles, I have designed AI-powered systems involving autonomous agents, LLM-driven enrichment and inference pipelines, knowledge retrieval, embeddings, vector search, semantic search, and context-aware automation. My work spans the full lifecycle of these systems, from data ingestion and retrieval pipelines to agent orchestration, APIs, infrastructure, observability, and user-facing applications.
I have also built and maintained high-scale distributed systems, including Python and Node.js microservices handling more than 2 million daily requests with 99.9% uptime. I place strong emphasis on reliability, performance, maintainability, security, and observability, using tools and practices such as Kubernetes, Docker, PostgreSQL, Kafka, Grafana, Kibana, and automated testing and CI/CD.
My technical experience includes Python, TypeScript/JavaScript, Node.js, React, FastAPI, Flask, Django, NestJS, PostgreSQL, Redis, Docker, Kubernetes, AWS, GCP, BigQuery, LLM APIs, LangChain, RAG, vector search, and agentic AI workflows. I am comfortable working across frontend, backend, data, infrastructure, and AI layers when the problem requires end-to-end ownership.
Currently concluding ongoing projects and seeking a long-term role with ownership, where I can contribute to building and scaling AI-native products.
Core focus:
- AI Systems Architecture (RAG, retrieval pipelines, agentic workflows)
- Distributed systems & microservices
- Production-grade LLM integrations & guardrails
- Observability, reliability & performance at scale
- Engineering leadership & delivery ownership
Leadership & Operating Style
- Managed and mentored 6â10 engineers
- Owned technical direction, delivery, and architecture decisions
- Contributed to hiring and performance feedback
Operating style:
- Async-first, low-meeting environment
- Strong documentation and ownership culture
- Focus on outcomes over activity
- Balanced roadmap: ~70% features / 30% tech debt
Work Experience
- Architected a centralized data warehouse (Google BigQuery) integrating Shopify Plus, Google Ads, and supplier APIs across a 1M+ SKU catalog, enabling unified data access and analytics at scale
- Built Python-based autonomous AI agents to dynamically adjust pricing and inventory tagging based on real-time sales signals and business rules.
- Designed and implemented LLM-powered enrichment and inference pipelines, transforming raw manufacturer data into structured, high-converting product content at scale.
- Owned end-to-end REST API integrations across Shopify, Google Merchant Center, and external supplier systems, ensuring data consistency, reliability, and maintainability.
- Developed agentic AI workflows combining retrieval, validation, deterministic output generation, and knowledge retrieval, improving automation reliability.
- Designed and implemented AI-powered knowledge retrieval using LLMs, embeddings, RAG, and vector search, enabling semantic search, intelligent navigation, and context-aware user experiences.
Contributed to scaling a $12M+ revenue engine through AI-driven pricing optimization, performance tooling, and scalable software engineering practices.
- Architected and maintained high-scale microservices (Python, Node.js) handling 2M+ daily requests
- Ensured 99.9% uptime via Kubernetes orchestration and optimized database performance
- Designed and implemented LLM-powered systems and agent-like workflows, enabling contextual reasoning, tool execution, and intelligent automation
- Built RAG pipelines combining structured + behavioral data for semantic retrieval and personalization
- Developed AI recommendation engine for real-time adaptive user interactions in production systems
- Engineered memory and context-aware systems for long-running conversational and behavioral personalization flows
- Built Python-based AI infrastructure for validation, monitoring, and feedback loops, improving reliability by 40%
- Implemented observability and evaluation pipelines (Grafana, Kibana) for tracking AI behavior and system performance
- Led end-to-end development of web and desktop applications using TypeScript, React, Node.js, and Electron, delivering scalable Express.js APIs, user-facing features, and AI-assisted developer tooling.
- Gained hands-on experience with Rust through policy primitives, code verification, and performance-oriented components, applying Rust's ownership and safety concepts while collaborating on systems requiring high reliability.
Technologies: Python, Node.js, TypeScript, Flask, Django, Express.js, Electron, PostgreSQL, React, Redux Toolkit, Rust, Vue 2, Docker, Kubernetes, Grafana, Kibana, OpenAI API, Anthropic Claude API, LangChain, RAG pipelines, vector search concepts, JWT, MSAL