Follow up decisions from meetings
Clear owner, date and next action. Then stay with it until it is closed.

Snevagten A/S · Executive Operations Associate · targeted portfolio
My recent background is technical purchasing and operations, but the common thread is coordination: keeping suppliers and internal teams aligned, making data usable, and keeping projects and open actions moving.
That is why this role makes sense: support the Senior Partner Success Manager, keep partner commitments and projects moving, protect data quality and qualify AI agents against real work.
What I understand the job to be
The role is broad, but the operating problem is simple: keep decisions, partners, projects and information moving while the business and AI team improve the tools around them.
Clear owner, date and next action. Then stay with it until it is closed.
Make commitments, changes, risks and next steps visible before they become bigger operating problems.
Check the source, resolve conflicts and give the manager something clear enough to act on quickly.
Test normal cases, missing or conflicting input and awkward exceptions before an agent becomes part of the operation.
On a steel order, I challenged the proposed timing with market and production data, gave my recommendation, respected the manager's final call and kept monitoring the risk afterward.
At A.K.S. I coordinated drawings, ERP/BOM/routing data, supplier and material readiness, production documentation, FAI/traceability and open actions across Engineering, Quality, Production, Purchasing and suppliers.
RFQ → comparison → negotiation → PO → confirmation → follow-up → shortage or delivery recovery. I looked beyond unit price to lead time, capacity, quality, inventory exposure and operational consequence.
How I would support the Senior Partner Success Manager
I am comfortable with the practical side of support: trackers, follow-up, process notes, work instructions and clear handovers. At A.K.S. I handled routine decisions independently and brought larger framework or higher-risk questions to the Head of Purchasing with the facts, options and consequences already clear.
Open decisions, partner exceptions, missing information and anything that needs escalation.
Projects, partner commitments, corrections and milestones that can disappear behind daily noise.
Manual checks, unclear handoffs or weak data points worth fixing rather than working around again.
Working style: if I can bring the facts, impact, recommendation and next action, I do not bring a manager a vague problem.
How I would qualify an AI agent for real work
Define what the agent may read, draft or suggest, where human approval is required and when it must stop. Then test it against real cases before wider use.
What is the user trying to get done? Which source is trusted? What may the agent read or suggest, and what must it never guess?
Routine input, missing information, conflicting information and unusual but valid situations.
Compare the output with the trusted source and operating context. High-impact actions stay with a person.
Show IT what failed, what should have happened, when the agent should hand over, and whether the revised behaviour now works.
I bring the business problem, approved sources, permission and approval boundaries, exceptions and realistic test cases. The development team owns the architecture; I help qualify whether the agent behaves safely and usefully in the operation.
Material arrived without required CoC/COO. I held it, kept Production informed, chased the supplier and moved to an approved alternative when the origin was not acceptable. I would use the same principle when an AI agent lacks a trusted source, reaches a high-impact exception or should hand control back to a person.
Practical workflow evidence
I created and built the full operating concept: service architecture, buyer journeys, audit methods, source-of-truth rules, agent-readiness, approval and fallback logic, multilingual content, practical tools, QA and deployment. I repeatedly tested pages, forms, journeys and outputs across EN/DA/FR and desktop/mobile. The result is a live system for turning messy business processes into controlled, testable AI-supported workflows and clear technical handovers.
I conceived and built the pilot from idea to working flow: separate journeys for candidates, companies and employment professionals; friction and source mapping; task-owner logic; review rules; 7/30/90 routes; consent and privacy; multilingual guidance; human review; and handover toward official routes such as VITAS. I deliberately removed automatic scoring and avoided collecting CPR numbers or diagnoses.
I built the digital operating layer around a live Valby business: service and pricing architecture, policies, booking logic, bilingual website, reviews and trust, local visibility, influencer booking codes and customer journeys across the site and Planway. I manage release QA, responsive fixes and consistency between service, price, policy and booking as the business changes.
View live siteEvidence from real operations
When capacity became the constraint, I prioritised critical demand, controlled commitments and coordinated alternatives. With recurring external-processing issues, I also tightened checks, communication and ownership instead of treating every case as new.
Capacity · priorities · alternatives · repeatable controlsThe IT Manager asked me directly to clean duplicate, obsolete, incorrect and ghost/orphaned records. During incoming customer PO/order and related commercial checks, I also identified conflicting cost/sales-price data affecting 100+ items, traced the issue and coordinated correction.
IT-requested cleanup · 2,000+ records · 100+ affected items · correctionFor about three weeks I covered work normally shared among three people, prioritising what could stop execution and keeping open actions and handovers clear as priorities changed.
~3 weeks · prioritisation · autonomy · handoverAt AOT / ATS Pacific I coordinated French and European customers, suppliers, confirmations, changes, complaints and urgent recovery by phone and email.
Customers · suppliers · phone/email · urgent recoveryFirst month
I would rather make one useful workflow reliable than arrive with ten AI ideas before I understand the job.
Understand the Senior Partner Success Manager’s recurring meetings, partner commitments, partner platform, trackers and decision points, then become a dependable super-user of Snevagten’s platform, Claude Workspace and the internal AI agents and tools already in use.
Carry actions, partner exceptions or project follow-up with owner, date and status visible so fewer points depend on the manager remembering or chasing them.
Start with one repeated partner or data-quality case chosen with the Senior Partner Success Manager, for example missing or conflicting information during onboarding or recurring operational follow-up. Map how information moves between office, operations and partner; define what an AI agent may check or prepare; set human approval and stop rules; log failures; and retest with IT before wider use.

Executive Operations + AI
Romaric ClementiCONTACT
Greater Copenhagen · EU citizen · French native · English fluent · Danish operational (PD2/B1+; SAP S/4HANA course completed fully in Danish).