# Sercan Esen

Sercan Esen builds AI tools for workplace safety at Intenseye and works with manufacturing customers to understand safety challenges on the factory floor.

- Profile: https://b2binfluencer.marketing/creators/sercan-esen/
- LinkedIn: https://www.linkedin.com/in/sercan-esen
- LinkedIn followers: 20,649
- Topics: artificial-intelligence, industrial-ai, manufacturing
- Based on 13 dated posts, latest 2026-09-30

## Summary

Sercan Esen is most useful for workplace safety AI field research and customer facing talks with manufacturing customers.

- Best for: Fits workplace safety technology campaigns needing manufacturing customer field research or a customer facing webinar speaker.
- Not for: Does not fit legacy EHS compliance and audit platform campaigns seeking public endorsement for a compliance suite.
- Check first: Confirm partnership format, rates, and whether audience includes EHS or plant leaders.

### Takeaways

- They argue safety is work design, not compliance monitoring, and light curtains cannot stop walking around.
- They say EHS, operations, and IT must align, and tools should support decisions on the floor.
- They report more than 35 internal AI agents shipped and 621 customer onsites since 2025.

## Main ideas

- **Trust, not demos, is the test of a safety vendor**: Esen argues that the more AI can do, the more customers want to know whether the vendor will be there, understand their world, and build with them. Their answer is presence: sitting in the same room, shaking hands, and returning repeatedly. In their World Economic Forum post, the same idea appears from the technology side: human-machine collaboration fails if frontline workers do not trust the systems or leaders quietly bypass their own safety rules.
- **Redesign the hazardous step rather than asking workers to try harder**: Esen says behavior is the real frontier of safety innovation. A light curtain can stop a machine, but it cannot stop someone from walking around it, so detection alone does not fix the problem. The better move, in their account, is to redesign the step so the machine takes on the risk and the worker stays out of harm's way. They report growing skepticism toward vague "proactive" and "predictive" claims and say the bar is rising toward engineered outcomes that remove risk at the source instead of adding an administrative layer or a reminder. Automation, they argue, should remove exposure, not people.
- **Plant visits are the research method, and the findings are specific**: Esen describes a repeating loop: visit a customer site, watch the work on the floor, fly home, build what is missing, repeat. The trip counts represent firsthand exposure, not desk research, and the conversations are grouped to show which function owns what. Their summary: EHS owns outcomes, operations owns reality, and IT scales it. When the three align, execution accelerates; when they do not, well-funded initiatives stall.
- **Builders over slide decks, measured in shipped internal agents**: Esen writes about building on planes between customer visits and late at night after days on the floor, and says better tools have cut some work from a month to a week. They report 35 or more internal AI agents live in production, some for teams serving customers and some for forward-deployed engineers in the field, and argue that agents do not replace builders but multiply them. Profiling work, in this telling, is done by people who show up on the floor and then sit down to build.
- **Safety data is operational data, so safety needs new roles and a strategy seat**: Esen argues that in modern factories, safety data is operational data. When leadership treats it as a core data stream, safety teams become workflow architects and frontline workers become partners to the technology, rather than rule followers who hear about it only when something goes wrong. From that shift, they predict new industrial jobs, not just new tools.
- **Compliance platforms have stalled; decisions matter more than dashboards**: Esen says compliance still matters, but many EHS compliance platforms have not evolved: adoption is low, manual work is high, and workflows sit far from the floor. Some vendors relabel their products with "AI" without operationalizing the data into daily work, leaving the outcome unchanged, they say. Their standard is adoption in practice: if a tool does not show up in daily operations, it becomes nice to have, and in manufacturing, nice to have gets cut. Dashboards are not the ask, they write; decisions are. The value lies in helping frontline leaders make the right call in the moment.

## Work to read

- [Contributing to the World Economic Forum’s framework on human–machine collaboration](https://www.linkedin.com/posts/sercan-esen_home-the-next-frontier-of-operations-activity-7401608229786464256-hWvn)
- [22,487 miles: five patterns from manufacturing site visits](https://www.linkedin.com/posts/sercan-esen_ehs-manufacturing-operations-activity-7409186117675266048-dbVI)
- [“I love building”: 35+ internal AI agents in production](https://www.linkedin.com/posts/sercan-esen_ehs-workplacesafety-ai-activity-7465770114865381376-CMDL)
- [621 customer onsites, and why showing up is the trust argument](https://www.linkedin.com/posts/sercan-esen_ehs-manufacturing-safety-activity-7505949402008592384-FL1Z)
- [“You don't have to wait for backflipping robots”](https://www.linkedin.com/posts/sercan-esen_you-dont-have-to-wait-for-backflipping-activity-7462541886793289728-fxDz)

## Brand relationships

- **Intenseye** (owned_business)
- **World Economic Forum** (disclosed_partnership)

## Partnerships

Sercan Esen's posts contain no sponsorship menu, media kit, rate card, or partnership inquiry page, so there is no published creator-partnership format to plan against. They show how Esen and Intenseye work with manufacturers: repeated customer onsite visits (one post counts 621 onsites completed since the start of 2025), an open invitation to request a plant walk by comment or direct message, and a contribution to the World Economic Forum's Global Framework on Human-Machine Collaboration. These are company-level sales and research motions, not documented sponsorship offers; the posts give no pricing, deliverables, exclusivity terms, or lead times. For a customer-facing webinar or field research collaboration, expect to negotiate terms directly and ask for a written scope covering audience, session format, data handling for any plant-floor research, and approval steps.

## Alternatives

- Joe Renz: Choose Joe Renz when the campaign focuses on EHS software adoption and change management rather than plant-floor safety technology, or when the host wants a practitioner-dialogue format. Renz's strongest post for this buyer decision argues that the most expensive safety program is the one nobody uses: six-figure EHS software sitting at 20 percent adoption a year after go-live, with dashboards that look fine while the floor still runs on paper forms and email incident reports. The stated cause is that implementation assumed people would adapt to the system instead of the system adapting to people. The proposed benchmark: the newest operator on the floor can use the system on day one without training. For a safety-technology vendor whose customers complain about low adoption of tools they already own, this is sharper and more specific criticism than Sercan Esen's broader claim that compliance platforms are being challenged. Renz also uses a two-way format. A question to EHS leaders about what a new EHS manager should ask in the first 30 days drew replies from Bradley York on whether corrective actions come out of the EHS budget, Michael Kary on asking everyone from the ground level to the CEO what gets in the way of making the safe choice, and Shannon Dickson Rand on who investigates incidents and who is accountable at the front line. If the host wants a webinar where the audience supplies the content and the speaker structures it live, Renz has demonstrated that method.
- Tim Regnier: Choose Tim Regnier when the host's customers work in warehousing, distribution and retail supply chain rather than factory safety, or when the panel asks how organizations prepare frontline work for AI execution. Regnier's argument for this buyer task follows a readiness sequence: Gartner predicts 40 percent of agentic AI projects will be canceled by the end of 2027 while MHI data shows 65 percent of supply chain leaders plan to invest. The stated reason projects fail is that organizations skip groundwork and jump to implementation. The proposed fix is to make work AI-readable: map how work actually gets done, connect planning and execution systems to frontline activity, capture the daily exceptions people handle, and define what successful completion looks like, with the frontline inside the intelligence loop rather than receiving instructions from it. A second post offers a concrete frontline-performance angle: fulfillment and distribution centers with 20,000 to 200,000 frontline workers gaining an average 4 percent quality performance improvement and hitting productivity goals weeks faster with new hires. For a webinar about AI in industrial operations, this is a more developed explanation of why deployments stall than a tooling demonstration.
- Yvonne Masters, PhD: Choose Yvonne Masters when the host's manufacturing customers work in food and nut processing, or when the campaign needs an accredited instructor and published co-author rather than a technology vendor voice. Masters approaches the same buyer decision—bringing outside expertise to a customer-facing session—from the food safety culture side. The posts show a Preventive Controls for Human Food v2.0 course taught for food manufacturing employees, a presentation on creating the code on food safety culture at the Peanut and Tree Nut Processors Association Operations and Technical Food Safety Forum, a small-group exercise in which participants built actionable ideas for one nut manufacturer, and co-authorship of the Culture of Food Safety chapter in the association's handbook for the safe handling and processing of nuts. Masters combines teaching and facilitation: they run a scenario with participants, then point them to written guidance and tools. That is the closest match in this set to a working session rather than a talk.

---
AI-assisted summary of public posts. Audience demographics are not available. Engagement appears only where we checked recent posts. Method: https://b2binfluencer.marketing/methodology/