I'm Amir Firouzmanesh, a senior software engineer (PhD) and founder of AandZ.tech, an independent AI & cloud engineering practice based in the UAE. Over the last decade I've helped teams at Microsoft, Amazon Web Services, Property Finder, and DataGardens ship real products—from AI-powered photo editing in Windows Photos to RDS SQL Server features, marketplace analytics, and subscription/credit systems behind large real-estate platforms.

These days I focus on AI integration and event-driven architectures: layering LLMs, RAG, agents, and image models onto existing systems without turning everything into a research project. I like working end-to-end: clarifying requirements, sketching architecture, building services and APIs, instrumenting them with metrics and traces, and leaving behind clear docs and simple mental models. Much of that thinking lives in open-source reference projects—such as Credit Ledger, AI Image Search / Feature Router, and Learning Path Designer—that I use as blueprints for client work.

How I can help

• AI & GenAI integration: LLM/RAG/agent features, evaluation loops, guardrails, and safe fallbacks to ensure production reliability and minimize hallucination risks.
• Distributed systems & microservices: event-driven designs (Kafka/Redpanda, RabbitMQ, Kinesis), CQRS, outbox, and idempotency to ensure data consistency and system resilience at scale.
• Credits, ledgers & subscriptions: double-entry ledgers, usage/entitlement flows, and incident-resistant invariants to prevent revenue leakage and ensure 100% financial accuracy.
• Architecture, mentoring & technical consulting: architecture reviews, design docs, code reviews, and short-term "fractional principal engineer" engagements to unblock teams and reduce long-term technical debt.

Technologies I work with

• Languages: Go, Python, TypeScript/JavaScript, Java, SQL
• Backend & AI: FastAPI, REST/JSON APIs, ONNX Runtime, LLM APIs, LangChain
• Frontend: React, Next.js, Tailwind CSS
• Data & storage: PostgreSQL, MySQL, SQL Server, MongoDB, Qdrant/pgvector
• Cloud & infra: AWS, Azure, Docker, Kubernetes
• Messaging & observability: Kafka/Redpanda, RabbitMQ, Kinesis, Grafana, Prometheus, Jaeger, Application Insights