03.07.2026 aktualisiert
100 % verfügbar
Freiberuflicher KI-Architekt & Software Engineer — GenAI von PoC bis Produktion
Hamburg, Deutschland
Nur Remote
Master of Science in IT Management and Consulting (ongoing)Über mich
Ich bringe GenAI-Systeme in Produktion — nicht nur in Demos. Zuletzt: Multi-Tenant-Transkriptions-SaaS (FastAPI, React, Azure Speech, K8s/Helm), Speech-to-Text-System für eine deutsche Rechtskanzlei, RAG-Plattformen für Unternehmensdokumente.
Skills
RAGGenAILLMAgile MethodologieAgile ModellierungKünstlerische EntwicklungPythonSoftware ArchitectureTypeScriptMulti-Agent SystemsFastAPIKubernetesDocker
Agentic AI & Multi-Agent Systems
Design and development of multi-agent systems using LangGraph, LangChain, and LangSmith, including the creation of LangGraph-based applications that transform meeting transcripts into structured AI use-case specifications and quality-gated user stories.
LLM Integration & Prompt Engineering
Integration of large language models via OpenAI API and Anthropic API, combined with advanced prompt engineering techniques and RAG pipeline development for real-world business automation use cases.
Web & Backend Development
Development of web and backend applications using Next.js, FastAPI, PostgreSQL, and REST APIs, with deployment supported by Docker, CI/CD pipelines, and Git-based version control.
Programming Languages
Proficiency in multiple programming languages including Python, JavaScript/TypeScript, C++, C#, Go, and Rust, enabling versatile software development across different domains.
RAG Pipelines
Construction of Retrieval-Augmented Generation pipelines using tools such as pgvector and WhisperX to enhance AI-driven information retrieval and processing workflows.
Multi-Agent Orchestration
Orchestration of complex multi-agent workflows with human-in-the-loop validation, targeting research publication at venues such as REFSQ, IEEE RE, and CHI.
Tooling & DevOps
Experience with modern development tooling including Git, CI/CD, Docker, GitHub Copilot, Claude Code, and Codex to streamline development and deployment processes.
Human-in-the-Loop Design
Application of human-in-the-loop design principles in AI systems to ensure quality-gated outputs and reliable validation in automated pipelines.
Design and development of multi-agent systems using LangGraph, LangChain, and LangSmith, including the creation of LangGraph-based applications that transform meeting transcripts into structured AI use-case specifications and quality-gated user stories.
LLM Integration & Prompt Engineering
Integration of large language models via OpenAI API and Anthropic API, combined with advanced prompt engineering techniques and RAG pipeline development for real-world business automation use cases.
Web & Backend Development
Development of web and backend applications using Next.js, FastAPI, PostgreSQL, and REST APIs, with deployment supported by Docker, CI/CD pipelines, and Git-based version control.
Programming Languages
Proficiency in multiple programming languages including Python, JavaScript/TypeScript, C++, C#, Go, and Rust, enabling versatile software development across different domains.
RAG Pipelines
Construction of Retrieval-Augmented Generation pipelines using tools such as pgvector and WhisperX to enhance AI-driven information retrieval and processing workflows.
Multi-Agent Orchestration
Orchestration of complex multi-agent workflows with human-in-the-loop validation, targeting research publication at venues such as REFSQ, IEEE RE, and CHI.
Tooling & DevOps
Experience with modern development tooling including Git, CI/CD, Docker, GitHub Copilot, Claude Code, and Codex to streamline development and deployment processes.
Human-in-the-Loop Design
Application of human-in-the-loop design principles in AI systems to ensure quality-gated outputs and reliable validation in automated pipelines.
Erfahrung
Berufserfahrung5 - 10 Jahre
Freelancer seit2026
Sprachen
ArabischMutterspracheDeutschverhandlungssicherEnglischverhandlungssicher
Projekthistorie
Research and prototyping of state-of-the-art GenAI and Agentic AI methods, evaluating cutting-edge techniques against real-world use cases. Designing TRANSACT: a LangGraph multi-agent system (Next.js, FastAPI, WhisperX, pgvector, LangSmith) that converts meeting transcripts into AI use-case ideation, MVP specifications, and quality-gated user stories with human-in-the-loop validation. Collaborating with academic and industry partners to advance AI-driven solutions, targeting publication at REFSQ, IEEE RE, and CHI.
Leading digital infrastructure projects including website development, social media automation, and strategic AI initiative rollout. Evaluating and integrating AI tools such as LLMs and code assistants to improve team workflows and content pipelines.
Founded an AI solutions company in Dubai; built custom AI agents for business process automation and digital marketing. Developed and deployed LLM-powered tools for client use cases, combining prompt engineering with API integrations. Managed influencer and brand strategy for a portfolio of clients across the GCC region.