Freelancer profile translated to English.

Description

I design, launch, and **scale AI products that create measurable value**. My playground: copilots, voice AI, data/ML products, and Enterprise packaging/pricing. Dual role hands-on (PRD, evaluation/guardrails, integrations) and executive (strategy, OKR, GTM, management). My obsession: adopted use cases, rising metrics (adoption, retention, ARR), and running operations (SLO, quality, latency).

How I help you

Clarify the problem and align it with your business KPIs (no technology for technology's sake).
Move fast from Scope→POC→MVP with proof (golden sets, scenario replays, A/B tests).
Industrialize: guardrails, observability, escalation playbooks, post-launch reviews.
Sell & deploy: AI narratives that speak to decision-makers, pricing/packaging, sales/CS enablement.
Manage multidisciplinary teams (PM, Data, Eng, Design) with a clear cadence.

What sets me apart

0→1→scale track record on 3 AI product lines, plus copilots/voice agents in production.
Data-driven culture (cohorts, churn drivers, instrumentation) AND execution (rituals, quality, SLO).
Ability to simplify AI for Enterprise sales cycles (targeted demos, quantified value proofs).

Types of projects & deliverables

AI Discovery→MVP (4–8 wks): PRD, user stories, offline/online evaluation, industrialization plan.
LLM & Voice agents: playbooks, guardrails, AHT/FCR dashboards, API integrations, workflow, and automatic evaluations
Interim CPO: OKR, roadmap, pricing/packaging, quality reviews & go-to-market.
Deliverables: PRD & design specs, evaluation scorecards, architecture/observability diagrams, GTM decks, production runbooks.

Talk to me about a business objective, and I'll translate it into a deliverable, measurable, and adopted AI product!

Industry field of expertise

Languages

English
Native or bilingual

Workplace preferences

Can work on-site

Paris (up to 50km)

Sunrise
AI Product Director
MEDICAL
August 2025 - Today (10 months)
Paris, France
Freelance AI Product Director in digital health, I manage the Scope → POC → MVP → Scale chain for clinical copilots and voice agents serving contact centers and sleep physicians, with cross-functional leadership (Ops, CC, Finance, Data, Eng, Design).
Agent Governance & Quality: offline golden sets, regular scenario replays, online guardrails, and continuous observability; monitoring intents, grounding, escalations, and AHT/FCR/CSAT KPIs.
Patient Ops & Stack: complete redesign of communication (irritant audit, standardized playbooks) and implementation of a stack (Front for omnichannel routing, SLA, analytics) integrated with EHR and telephony.
Compliance & Explainability: traceability, consents, audit logs; formalized clinical explainability requirements.
Delivery Cadence: weekly demos, quality gates, post-launch reviews (latency/relevance/security); tool-assisted triage and remediation.
Impact: ↓ AHT, ↑ FCR, scaled through continuous self-evaluations
Artificial Intelligence Guardrails and Agent Evaluation Generative AI Product Discovery Observability
MyTraffic
AI Product Director
SOFTWARE PUBLISHING
June 2022 - June 2025 (3 years)
Paris, France
Interim CPO and AI Product Director mandates to design, launch, and scale 3 AI product lines in location intelligence (from 0→1→scale). AI/data strategy linked to impact metrics (adoption, retention, revenue contribution) and GTM (packaging/pricing/enablement). Management: 4 Senior PMs + 1 Senior Designer; alignment of Engineering / Data / Design through OKRs, rapid decision-making rituals (Discovery → precise PRDs → Design specs → Ship), and post-launch reviews (quality, latency, guardrails).
GTM & Monetization: offering definition, pricing models, positioning; sales enablement (decks, battlecards, objection handling).
Hands-on Sales Support: joint enterprise pitches, advanced discovery, narratives that make AI buyable, targeted demos, and quantified value proofs.
Value Measurement: product instrumentation, adoption/retention/revenue dashboards, cohort analyses & churn drivers.
Ops & Quality: release cadence, relevance/latency SLOs, triage/backlog, observability, and guardrails (post-mortems, remediation playbooks).
Adoption & Training: workshops, customer playbooks, external content to accelerate onboarding and secure usage.
Impact: newly monetized AI offerings, shortened time-to-market, strengthened pipeline, sustainable execution discipline (cadence, quality, metrics).
AI Product Management Product Leadership Product Roadmap Prioritization (RICE, MoSCoW, ...) Sales Enablement
Omdena
Machine Learning Engineer
September 2019 - August 2020 (11 months)
ML Engineer within an international team, I designed and trained a U-Net model (Keras/TensorFlow) on satellite imagery (Google Earth Engine) to map micro-solar grids in Nigeria. I managed the end-to-end chain: data ingestion & cleaning, geospatial pre-processing, data augmentation, distributed training, and evaluation (IoU, F1) before experimental deployment.
Beyond modeling, I structured dataset traceability (dataset/experiment versioning), documented reproducible pipelines, and led code reviews to ensure quality and maintainability.
Impact: significant improvement in solar infrastructure detection, providing decision-support maps for deployment. Work conducted with an ethical approach (bias, limitations, uncertainty transparency) and MLOps reflexes (datasheets, logs/monitoring) that now inform my end-to-end AI product management (quality, observability, reproducibility).
Computer Vision Geospatial Data AI Product Management Data Product Artificial Intelligence