Florent
Salettes
Product Manager — Data & AI
30-second read
The static dashboard is over. Data delivery becomes dynamic.
Product Manager, Data & AI in B2B SaaS, from offer strategy down to SQL. Reporting to the CPO at Opsealog: 350 vessels, 8 oil-major clients.
Four things to take away
Track record
Profile
Product Manager, Data & AI, in B2B SaaS. Over six years of product ownership, currently reporting to the CPO, owning the strategy of a whole data and AI scope end to end: discovery, value framing, architecture trade-offs, production, pricing. Equally at ease with a business case and with a technical arbitration. I turn scattered signals into structured product decisions: collection, qualification, scoring, action. My conviction: the static dashboard is over — data delivery becomes dynamic, a targeted recommendation for every user, application or AI agent instead of a table to interpret. I ship AI-powered products, I prototype what I used to only specify, and I move things without hierarchical authority. French native, English C1.
Skills
Product
Product vision and strategy, product discovery, customer interviews, Voice of Customer, opportunity qualification, prototyping and experimentation, roadmap, value-based prioritization, business cases and ROI, build/buy trade-offs, pricing and packaging, go-to-market, sales enablement, end-to-end delivery, Agile Scrum.
Data and AI
Data product management, data platform, medallion architecture, information architecture, data governance, data contracts, data quality, semantic layer, data sharing and monetization, API product design, data readiness for AI, AI product governance and evaluation (shadow testing, human-in-the-loop), AI agents, MCP.
Tools
French native, English C1.
Experience
Opsealog
B2B SaaS, offshore maritime fleet performance — Marseille | Reporting to the CPO | Working with a 5-person data squad | 350 vessels, 8 clients (oil majors)
Contractual title: Product Owner Data & Analytics
- Product strategy and discovery — Designed and rolled out the framework feeding the roadmap: 7 collection channels, around 200 customer and business signals broken down atomically, weighted into insights, converted into some twenty qualified opportunities arbitrated in an Offer Committee. Now the single entry point to the roadmap.
- API-first data product and monetization — Structured and launched a data product exposing around ten datasets and REST endpoints: API contracts, versioning, access model, governance. Set up the developer portal and onboarding path, machine-to-machine authentication and scoped access, usage and cost monitoring. Built the subscription pricing grid indexed on managed assets, applied to new contracts and renewals, with client documentation and sales material. Two clients consume the API in production, around 500 requests per day per client. Data also exposed through an MCP server for consumption by AI agents.
- AI agent for self-service analytics — Owned an agent answering natural-language questions on operational data and returning text, tables or charts through SQL generation and execution. Defined capabilities, prioritized use cases, ran business validation, tracked product metrics: 95-100% response rate, 2-3 minute response time, classified SQL errors, monitored cost. Deployed to production and deliberately kept in controlled internal testing through a shadow-testing protocol preserving the human baseline, on a cross-functional panel of 10 users.
- Data governance, semantic layer and information architecture — Defined the platform’s data governance model (medallion architecture, ongoing platform migration) with business teams. Functional owner of inbound and outbound data contracts, business definitions and their lifecycle, and quality standards. Building an AI-ready semantic layer, released progressively, feeding MCP exposure and the analytics agent.
- Business automation engine — Designed and built alone an engine for qualification, scoring and action triggering (SQL, action queue, workflow automation), automating contacts, follow-ups and escalations. Used in real conditions by business teams across the whole fleet (350 vessels), with several checks a day over 3 months. Later taken over and adapted for other use cases.
- From lookup tables to predictive models — Wrote the PRD and target architecture for rebuilding fuel consumption forecasting models, triggered by a client incident questioning their reliability. Used more than 10 years of historical data to build the dataset and select variables, set a 90% reliability target and aligned top management. Run with the 5-person data squad.
- Embedded sensor data integration — Produced the vendor benchmark and defined the integration model carried into an RFP response: 25 potential vessels, commercial offer modeled with the business team.
- Analytics delivery — Built and maintained around ten analytical datasets used in self-service by internal teams and clients. Ran the data roadmap in Scrum: trade-offs with product, tech, support and business teams, operational and contractual management of external data suppliers.
Teamwill Consulting
IT consulting firm
- Digital Project Manager — Ma French Bank assignment (04/2019 – 10/2020): launched the revolving credit offer on the business side, from requirements to deliverable sign-off. Subscription and drawdown journeys, functional specifications, user acceptance testing.
- Product Owner — GE Money Bank assignment (11/2017 – 01/2019): ran the roadmap and product backlog in Scrum for a securitization application: new features, application diversification, new operation types.
- Functional Project Manager — GE Money Bank assignment (12/2015 – 11/2017): took a securitization application live: full functional design (workshops, data models, interfaces), specifications, test management through go-live. Fixed-price waterfall project.
- Business Analyst — GE Money Bank assignment (01/2015 – 08/2015): formalized business needs, the functional solution and inter-application data flows during a pre-sales phase.
Side projects
Generic reimplementation of a business workflow automation engine (trigger, qualification, scoring, action) and an AI-assisted product opportunity qualification engine: structured multi-source extraction, scoring, explicit human validation before any write. Both built hands-on with AI coding tools.
Education
Generalist Engineering degree (MSc) — Arts et Métiers ParisTech (ENSAM), Paris