About Samuel
AI Engineer | Data Scientist | Digital Ads Specialist
I help companies automate smarter, sell faster, and grow bigger — powered by AI & data.
⸻
💼 What I do:
I build custom AI bots, automate repetitive workflows, and optimize digital ad campaigns to generate leads, save time, and drive revenue.
⸻
📊 Real Results:
• 💸 Managed €400,000+ in ad spend across Google, Facebook, LinkedIn & TikTok
👉 Delivered +1.4M€ in revenue, reduced CPL by 30%, boosted lead quality across all funnels
• 🤖 Built 15+ custom GPT-powered bots (sales, legal, HR, customer support, LinkedIn automation)
👉 Examples:
• Legal AI bot processing 5,000+ docs/month → saved 40% review time
• LinkedIn bot generating 100+ qualified meetings/month
• Multilingual support bot with 90%+ accuracy in real-time replies
• 📈 Deployed predictive models for lead scoring, customer segmentation, and campaign optimization
👉 Clients saved 2,500+ hours/year and cut human error by over 80%
⸻
🧰 Tools I Master:
Python, OpenAI (GPT-4), LangChain, Zapier, Make, SQL, Meta Ads, Google Ads, LinkedIn Ads, TikTok Ads, Notion API, Pinecone, Power BI, Streamlit, FastAPI
⸻
✅ Let’s turn your business into an AI-augmented machine.
Whether you’re a startup, agency, or SME — I’ll bring the automation and performance your team is missing.
French
Native or bilingual
Hebrew
Native or bilingual
English
Native or bilingual
Remote only
Primarily works remotely
Experience
- LeoData scientistCIVIL ENGINEERINGJanuary 2023 - February 2024 (1 year and 1 month)Paris, FranceRole: Data Scientist – Leo (Energy Renovation Sector)At Leo, a company specialized in energy-efficient home renovations, I worked as a data scientist for one year with a mission to turn operational data into actionable insights for sales and strategic growth.🔍 Key Responsibilities• Lead Conversion Prediction:Developed scoring models to predict the conversion likelihood of incoming leads using behavioral data (response time, call history, lead source).👉 Result: +27% increase in conversion rates for the sales team.• Advanced Customer Segmentation:Used clustering techniques (K-Means, DBSCAN) to identify key customer profiles based on location, motivation (ecological vs economic), and buying behavior.👉 Enabled targeted marketing campaigns with higher ROI.• Ad Campaign Optimization:Built dashboards and recommendation models to guide the marketing team on which acquisition channels (Facebook Ads, Google Ads, etc.) to prioritize.👉 Achieved a 30% reduction in cost per lead within 3 months.• Automated Performance Reporting:Automated weekly performance reports for management and field teams using Python, SQL, and Power BI — tracking KPIs, delays, anomalies, and lead progress.• Strategic Decision Support:Delivered actionable insights for project prioritization, installer evaluation, and fraud detection across customer and lead databases.🧰 Tech Stack• Python (Pandas, Scikit-learn, XGBoost)• SQL (BigQuery & PostgreSQL)• Power BI / Tableau• Zapier & Notion for internal process automation
- SV DATA CONSULTINGDigital MarketingE-COMMERCEJanuary 2019 - Today (7 years and 5 months)Paris, FranceAs a freelance consultant, I worked with companies in the energy renovation, e-commerce, and B2B services sectors. I combined data science expertise with high-performance digital advertising strategies, managing over €400,000 in advertising spend across multiple platforms.⸻📣 Digital Marketing Achievements• Managed large-scale ad budgets:Oversaw and optimized campaigns across:• Google Ads: €150,000 spent → average CPL reduced from €52 to €34• Facebook Ads (Meta): €180,000 spent → +45% increase in qualified leads• LinkedIn Ads: €40,000 spent → targeting B2B decision-makers with +300% ROAS in niche sectors• TikTok Ads: €35,000 spent → low-cost awareness campaigns with < €0.03 per view• ROI-Focused Strategy:Developed and executed funnel-based strategies (TOFU–MOFU–BOFU), combining creative testing, behavioral segmentation, and retargeting.👉 Delivered up to €1.2M in revenue generated across all client portfolios in less than 12 months.• Landing page & CRO optimization:Used heatmaps, A/B tests, and scroll-depth analysis to boost page performance.👉 Achieved a +23% lift in landing page conversion rate.
- GoogleData ScientistJanuary 2022 - January 2023 (1 year)Led the redesign of core ranking algorithms for Google Search, improving click-through rates (CTR) by +12.6% across key verticals (News, Shopping).Developed and deployed a BERT-based classification model to better interpret ambiguous queries, reducing user bounce rate by -9.3% in A/B testing.Collaborated with the Ads team to implement real-time bidding optimization, increasing ad revenue by $37M/year through better prediction of user intent.Created automated pipelines using TensorFlow, BigQuery, and Airflow, reducing model deployment time from 3 weeks to under 5 days.Conducted extensive causal inference analyses to measure the impact of UI changes on user behavior using DoWhy and Google CausalImpact.Mentored 4 junior data scientists and collaborated with cross-functional teams including UX researchers, software engineers, and product managers.Presented findings to senior leadership, influencing roadmap prioritization for Search Intelligence in EMEA region.
Recommendations
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Ingenieur Data SciencePolytechnique2022Specialization: Applied Mathematics and Computer Science Graduated: 2022 Rigorous multidisciplinary training in mathematics, physics, and computer science, complemented by courses in economics and innovation. Completed competitive coursework in machine learning, statistics, algorithm design, and data analysis. Conducted a 6-month research internship on optimization techniques for large-scale machine learning models. Member of the Data Science student club; participated in multiple hackathons and industry case studies. Ranked in the top 10% of the program (if applicable).