Paul-Darius Sarmadi

data scientist

Paris, France

  • 48.8546
  • 2.3477
Nouveau
Proposer un projet

Vérifications

Influence

Rejoignez la communauté Malt pour découvrir le profil complet de Paul-Darius Sarmadi
Rejoignez la communauté Malt pour découvrir le profil complet de Paul-Darius Sarmadi

Découvrez ce que recherche Paul-Darius

Déplacement
Peut faire la totalité d'une mission dans vos locaux
Zone d'activité
Recherche des missions à Paris et 10km autour
Durée de mission
Recherche des missions ≤ 1 semaine, ≤ 1 mois, entre 1 et 3 mois, entre 3 et 6 mois

Paul-Darius en quelques mots

Je travaille en tant que data scientist. Ce titre très large signifie plus spécifiquement dans mon cas que je me concentre sur :
- la création de beaux jeux de données, afin de construire des modèles à l'état-de-l'art, en utilisant les dernières technos.
- le déploiement de ces modèles en production.
- la conception de workflows destinés à faciliter la coopération entre d'une part data scientists au mode de travail parfois chaotique et d'autre part, équipes de développement agiles plus "traditionnels".

J'aime particulièrement les pipelines écrites en Python (bien que j'admire aussi les librairies R), et les releases en prod via AWS.

== In English ==

I work as a data scientist. This very broad title in my case specifically means I focus on:
- creating great datasets to build state-of-the-art models using state-of-the-art technologies.
- deploying these models in production.
- building smart workflows that make R&D oriented data scientists chaotic flows work efficiently with more "traditional" agile development teams.

I enjoy data pipelines written in Python (I also admire R's fancy libraries) and production releases with AWS.

Expériences

septembre 2017 - août 2019 | Paris, France

Edition de logiciels

Concord

Machine Learning Engineer

I - Developed ML (NLP) based product features from the ground up i.e. from data acquisition to production-ready API.

- Goal: Automatically extract key data from scanned contract documents - main dates and durations, parties.
- Data Acquisition: Managed remote teams to create several datasets of labeled contracts used to train ML models.
- Cleaning: detection and correction of labels, classes balancing.
- Machine learning: Built a python package that:
1) Extracts a collection of potential key data from a contracts database.
2) Trains binary classifiers (Random Forest) to select the right value from that collection. For example train a model to choose a contract’s effect date among all its detected dates on a large database of contracts.
3) Apply the “data extraction + trained model” pipeline on customers
documents.

II) Microservice delivering ML
Built an AWS based pipeline with terraform that delivers machine learning predictions as an API and stores predictions on CockroachDB.

- POC involved deep learning with keras.
- Built a project workflow to conciliate data scientists’ R&D processes with agile software development.
- Built a clean data science environment based on versioning of code, experiments with github, and data and models with DVC and Amazon S3.
septembre 2016 - août 2017 | Paris, France

Conseil & audit

Weave

Data Scientist

Data wrangling, visualization, and machine learning for several companies -mainly banks. Worked mostly on recommendation systems.
Main tools involved during missions:
- R: tidyverse, RMarkdown, RShiny.
- Python: pandas, scikit-learn, Flask with PostgreSQL.
- Algorithms used for machine learning: Linear Regressions, CART, Random Forest or XGBoost most often.
janvier 2016 - juin 2016 | Bangkok, Thaïlande

Centres de recherche

Asian Institute of Technology Vision and Graphics Lab

Deep Learning Research Study

Worked under the supervision of Matthew Dailey at AIT Vision and Graphics Lab. Created a deep learning software with Caffe. Given a database of surveillance videos and images of a person, the software tells in which videos and at which time the person appears. Purpose: Detection of criminals.

2 recommandations externes

Formations

Langues

charter modal image

Notre succès est une oeuvre collective

Contribuez à cette réussite et au sérieux de la communauté en signant la charte du freelance Malt.

Signer la charte