I help experts and companies with AI automation and brand positioning. With a multidisciplinary engineering background, I have spent more than twenty years building production, quality and operations systems in automotive, textiles and advanced technology. From the automotive supply chain to advanced technology startups, and from Kocaeli to Guangzhou and Dhaka.
From material to machine, from line to factory: turning work into systems you can measure. That is what I have done for twenty years.
Today I am a core founding developer of INVAMAR A.Ş., an advanced technology and Internet of Things (IoT) startup in Gebze, and I am preparing to formally take up the role of Business Director. I led the company's conversion into a joint-stock company, set up its corporate governance structure and prepared it for investment. In finance, human resources, operations and business development, I set up a way of working that runs on AI agents and automation. I am now preparing the company for the EU and US markets, and I am also leading its shift from hardware sales to subscription-based revenue models.
Before that, I managed production, R&D and quality units across several plants at Assan Hanil (Kibar Holding), TB Sewtech (Toyota Boshoku) and AB Apparels. On Hyundai and Ford vehicle programs, I took parts from the quoting stage to zero-defect series production. I built a factory from scratch in Bangladesh and led a quality assurance team in China.
From 2013 to today. Before that, I worked as an engineer and manager in textile and automotive production.
I led the startup's conversion into a joint-stock company and set up a management, reporting and control structure ready for the investment process. I designed a way of working based on generative AI across finance, human resources, operations and business development. I run the grant and investment processes, the preparation for entering the EU and US markets (GDPR, CE, FDA) and the shift to subscription-based revenue models (SaaS, DaaS, MLaaS).
Next-Generation Mobility and Program Lead 2023–2024
At this joint venture of Kibar Holding and Hanil E-HWA, I managed the development and production readiness of high-volume interior trim parts supplied to automakers. By monitoring machine data, I raised line efficiency (OEE) by 15% in multi-shift production. By planning material needs in advance, I cut annual raw material waste by 12%. I organized digital skills training for more than 100 employees.
Process and Industrialization Lead 2021–2023
On Hyundai and Ford vehicle programs, I managed the transition of parts from the quoting stage (RFQ) to zero-defect series production (SOP). Before the lines went live, I used FMEA studies to identify and prevent more than 20 high-risk failure modes. I carried out more than 30 supplier audits for APQP and PPAP requirements.
On seat cover production lines, I applied Kaizen, Kanban, 5S and the Toyota Production System and cut production lead time by 20%. I took part in preparing for ISO/TS 16949 audits and kept labor-intensive lines at zero nonconformities.
In Dhaka, I built a factory with a budget of more than $2 million from scratch, on schedule. Before that, I managed quality assurance at plants in Guangzhou and cut international delivery times by 15%.
Speaker profile: Marmara Urban Forum · LinkedIn
Mission
Bring the trust earned in the field online; hand repetitive work to AI agents while decisions stay with people.
Vision
The system, the knowledge and the habits build up on your side, and keep working after I step back.
Philosophy
A principle from lean manufacturing: automating without removing waste only speeds up the problem. Narrow scope, owned results.
Working with an agency often means long processes, complicated approvals and big budgets. I work in a boutique way: I respect your time, and I build the work that makes experts like you visible at a scale that fits your budget.
| Typical agency | Boutique | |
|---|---|---|
| Contact | An account manager and go-betweens | The person who does the work, directly |
| Process | Long meetings, layers of approval | Short calls, clear steps |
| Budget | Big team, big invoice | Work scaled to your budget |
| Relationship | A project that ends at delivery | A long-term working partnership |
How I work
The same skeleton runs in every engagement; the content follows your priorities. You decide whether to continue at the end of each step. Large companies call this way of working Forward Deployed Engineer (FDE): the engineer who builds where the work happens.
In the first call we define the problem underneath the request, not the request itself. Most gains start here: no longer solving the wrong problem.
For experts, how you appear today and the evidence you have; for companies, a 1–2 day on-site review. The process as it actually runs, not as documented.
The baseline, target and acceptance criteria are written down. If there is no measurement, we set one up first.
Unnecessary approvals, double entry and waiting are removed. Only then is an agent set up.
The agent starts in shadow mode or at suggestion level. It is measured every week, and every week there is visible output.
An internal owner is trained; the prompt library, usage rules and accounts stay with you. When I step back, the work keeps running.
Technology philosophy
A badly chosen tool can break a well-designed flow. I choose tools by these four rules:
Values
These are not slogans; they are rules that show up in the proposal, the contract and every step of the work.
If the data is not enough or the target is not realistic, I say so up front. I do not write made-up figures or inflated claims.
The person you talk to is the person doing the work. No go-betweens, account managers or subcontractors.
At the start of every engagement we write down what we expect on which metric. Results are measured, not claimed.
The agent prepares, you approve. Nothing is published or sent to a customer without your approval.
The system, the documents and the accounts stay with you. I do not accept ways of working that create dependency.
Client information is not used even as a reference without written permission. I sign an NDA before work starts.
Boundaries
Knowing what a consultant does not do matters as much as knowing what they do. I don't:
I provide operations consulting and fractional COO support to manufacturing and technology companies. I set up AI agents for your team's daily work and teach your team how to work with them. We can work by project or on a monthly basis.
A management rhythm that keeps a growing company from coming apart.
Making losses on the shop floor visible and reducing them step by step.
From prototype to zero-defect series production, on schedule.
Systems that hand repetitive work to agents and keep decisions with the team.
Data that flows in one place, from quote to shipment.
Products that reach EU and US markets with their paperwork ready.
“AI that is not used in the field has no value. Don't debate what you can measure; measure it.”
This page was last reviewed on .
