Digitalisation and automation specialist with more than 20 years of experience in how companies really work — and how those workflows can be automated. Twenty years of responsibility for commerce processes (B2B and B2C shops, marketplaces, product data, interfaces, warehouse and ordering processes) taught me the reality behind the processes that are automated with AI today: my own PIM systems in four companies, from the first ERP in 2001 to the AI-assisted generation in 2024. The focus was always the same: recognise recurring workflows, reduce manual work, centralise data.
I am currently extending this experience with a full-time AI Engineering programme and combining classic process automation with modern artificial intelligence — AI agents and assistant systems that I run in parallel in my own projects every day.
Roles in which more than 20 years of commerce and process responsibility come together with the AI work — each backed by a project. Based in Marl (NRW, Ruhr area, Germany); remote work is familiar.
Recognise recurring workflows and automate them with AI agents — from the trigger through model selection to checking whether the result actually holds. Multi-agent architectures across Claude, GPT and Gemini models, self-operated rather than presented.
Evidence: LISA has been self-operated daily since 03/2025: four machines with roles, 1,600 evidenced facts and 21,000 searchable memories.
Apply AI where it actually carries weight in daily business: assess the use case, align it with the departments, integrate it into existing systems instead of placing it beside them. Twenty years of commerce processes taught me where such rollouts fail.
Evidence: In PIM 3 (ITL, 2023–2025) AI product text generation was integrated productively via the ChatGPT API — aligned with purchasing, sales and logistics.
Own the target picture, requirements and roadmap of a technical product and prioritise it in expansion stages — close to development and to the business at the same time, with my own hands in the code.
Evidence: Product ownership of PIM 3 at ITL: target picture, requirements and roadmap developed and prioritised iteratively in expansion stages; at Askania, PIM modules for 3 departments rolled out in stages.
Work out from stock levels, sales and system breaks the question the business actually has — and translate it into analyses somebody can base a decision on.
Evidence: A.O.S. condenses stock analysis, route planning and notification systems into order proposals for 24 branches and 30,000 articles; the transfer analytics made unsold branch stock visible and turnable into revenue.
Remove paper and manual work from workflows: record the process, find the break points, rebuild it digitally — and measure the outcome, not the project plan.
Evidence: PickByScan replaced paper-based picking: shipping throughput +35 %, wrong shipments detected before dispatch. The PIM interface to JTL-Wawi eliminated transfer errors and manual rework.
Build solutions on generative language models: prompt and context design, API integration, and a rationale that can be read back — traceability instead of a black-box score.
Evidence: job4u evaluates job adverts in full text against an applicant profile: 870 adverts analysed, condensed to 141 solid matches — with traceable reasoning instead of a black-box score.
“Evidence” here means: the figure is on the linked project page and comes from the company where the system actually ran.
Focus: Product ownership · product data · interfaces · Magento · JTL · marketplaces · automation
Focus: Shop architecture · Magento · JTL · servers & databases · payment methods · shipping processes
Focus: Barcode & scanner · real-time warehouse routes · DHL connection · error rates · process design
Personal AI assistant built on the Jarvis principle
Self-operated daily since 03/2025: one assistant across four machines that knows my documents, remembers decisions and keeps every project running as …
Take a photo · AI recognises it · know the value instantly (app + Chrome extension)
Collection manager with value tracker as a PWA: one photo is enough — the AI recognises brand, model and condition, writes the DE/EN sales text and …
AI analysis of job postings
Evaluates job postings in full text against an applicant profile — with a traceable rationale instead of a black-box score. 870 postings analysed …
Indoor positioning via FRITZ!Box WiFi (open source)
Open-source tool that detects which room or floor someone is on in their own home, using the WiFi signal strength of multiple FRITZ!Boxes — for Home …
Development and daily operation of my own AI systems (LISA, Check4u, job4u, TA4U): multi-agent architectures across Claude, GPT and Gemini models — agentic …
Product ownership PIM 3: target picture, requirements and roadmap developed and prioritised iteratively in expansion stages — aligned with purchasing, sales …
Responsible for Magento 2 multi-channel shops (B2B, B2C, Amazon, eBay, Etsy) with more than 15,000 products
Magento shops (B2B/B2C) and marketplaces for a retail company with 24 branches and 30,000 articles (more than 8,000 actively online)
Accompanied two takeovers during live operation, from a mid-sized company into an international group; ERP interfaces between Sage Office Line and Intershop …
Set-up of online shops and eBay structures; additional responsibility for purchasing, sales and assortment
Set-up of online shops and eBay structures
Led a team of 14 employees; developed an in-house ERP including PIM 1.0 for more than 10,000 products; web applications for 23 subsidiaries; MDE barcode system …
Project and team coordination (6 developers) already during training; leadership responsibility after the takeover
Master School Institute of Technology (MSIT) · Remote · 09/2025 – 11/2026
Python · language models · prompt engineering · API integration · AI applications · deployment
More about the programmeIn 2025 it became clear to me: the tools I had built myself for two decades – product data maintenance, text automation, order proposals, warehouse processes – can now be created much faster and more flexibly with language models. I did not see that as competition but as an opportunity: whoever knows the workflows that are to be automated can achieve more with AI than ever before.
That is why I chose a deliberate fresh start: a full-time AI Engineering programme since 09/2025 and, in parallel, my own AI systems such as LISA, Check4u, job4u and TA4U in daily operation. My role stays the same as in more than 20 years of e-commerce – the bridge between business departments and technology – only with a new and much faster toolbox.
Dream big. Believe firmly in what you set out to do. The best AI builds nothing without your vision – but everything with it.
— Murat Danis
Away from code and commerce I am a passionate photographer – products, events or weddings. The camera is my creative outlet.
As a child of the C64/Amiga era I learned to make a lot out of little technology. That mentality still shapes me: pragmatic, resource-efficient, solution-oriented.
Currently I am fascinated by robotics (Arduino, Raspberry Pi), AI systems and everything that can be “brought to life”.
Murat Danis is a digitalisation and automation specialist from Marl (NRW, Germany) with more than 20 years of e-commerce experience: most recently Head of E-Commerce & Development at ITL GmbH, Pracht Creatives Hobby GmbH and Askania GmbH. Since 09/2025 he has been in a full-time AI Engineering programme (MSIT) and runs his own AI systems (LISA, Check4u, job4u, TA4U).
Four generations of his own PIM systems (2001–2025, most recently PIM 3 with ChatGPT API), several complete online shops from 0 to 100 (Magento, JTL, Amazon/eBay), PickByScan (paperless picking, shipping +35 %), an order proposal system (AOS), branch stock / transfer analytics and, since 2025, the AI systems LISA, Check4u, job4u and TA4U.
Yes – open to a permanent position, remote or in the Ruhr area. Currently in the full-time AI Engineering programme (until 11/2026); own AI systems run in daily operation in parallel. Based in Marl (NRW, Germany).
Turkish (native), German (business fluent, working language), English (basic with technical reading comprehension, AI-assisted in daily use).
Magento, JTL-Wawi, Amazon/eBay/Etsy, own PIM systems, Python, PHP, SQL, REST APIs; AI: LLM integration (ChatGPT, Claude, Gemini), Claude Code, OpenAI Codex, prompt engineering, multi-agent systems.
If you have a project in which data, processes or automation play a role: send me a short note about what it is – I will answer pragmatically and solution-oriented.
