Machine Learning Engineer (F/H/NB)

About

Ipsos Synthesio is a leading marketing SAAS company. Our technology powered with cutting-edge ML models allows global brands to make business decisions based on online consumer feedback. Synthesio centralizes public data from many providers, such as posts on social networks, press articles, forums, search data, and more. After the collection phase, Synthesio enriches it through AI and provides intelligence in many visualizations and APIs.

You want to join a deep tech and growing company with at-scale generative artificial intelligence applications, we are hiring candidates!

Lastly, at Synthesio, we have a strong tech culture. This means the teams have an important role. Each project team (“squad”) is empowered, owns together the success of the sprint, and chooses their deliverables and their timelines.

Our department

The Synthesio product engineering department is focused on data provisioning, enrichment, and exploitation.

Synthesio crawls numerous public consumer data footprints including social, reviews and ratings, survey, search, and press data, representing an average of 70M new documents per day. Each document is then analyzed and enriched using custom NLP and image recognition models developed by our ML engineers. All this data is stored in databases and can be accessed by our customers via our dashboarding solution or our APIs. Within dashboards, we harness users with the latest technologies, like Generative AI, to streamline their analysis.

ML stack Python, Pytorch, Huggingface libraries, sklearn, numpy, pandas, DVC, MLflow Serving: Sanic, Triton Inference server

Our platform

Our platform uses a microservice architecture containing 180+ microservices, most of them written in Go and using Kafka or HTTP to communicate.

The hardware infrastructure is composed of hundreds of rented bare metal servers running Debian. A big part of it is dedicated to storage, including:

1 PB+ Elasticsearch, 750TB+ MySQL clusters, 150 TB+ Scylla DB, and 250TB+ of Kafka.

Frontend side, we have two main frontend applications (built with React) and an in-house Design System.

Our ML microservices run on multi-GPUs machines

Job Description

Key Responsibilities: 

  • Develop and implement machine learning models using state-of-the-art algorithms in areas such as NLP, computer vision, and generative AI. 

  • Improve existing machine learning systems, e.g. automated sentiment & emotion analysis, named entity recognition, logo detection, scene recognition, topic modeling, etc. 

  • Perform statistical analysis and tune using test results. 

  • Extend the MLOps pipelines to improve data and model delivery speed and quality 

  • Keep abreast of developments in the field. 

  • Collaborate with the rest of the team during team sync meetings 

  • Write documentation and review merge requests

Preferred Experience

Technical skills 

  • Proven experience as a Machine Learning Engineer or similar role especially in NLP 

  • Strong background in computer science and software engineering: data structures, algorithms, object-oriented programming, code writing, and reviewing. Deep knowledge of machine learning algorithms (supervised, unsupervised, deep learning, transformers, ...), probability and statistics. 

  • Analytical and problem-solving skills 

  • Good written and oral communication skills. 

  • Ability to work in a team. 

Soft Skills 

  • Rigor and strong appetite for software quality 

  • Interested in manipulating dozens of microservices in data processing pipelines handling billions of documents 

  • Passion to discuss and explain technical choices 

  • Can-do attitude 

  • Good communicator, self-starter, and collaborative enthusiast 

  • Interested in understanding user needs 

  • Independent, self-organizing, and able to prioritize multiple complex assignments 

  • Interested in multicultural companies 

  • Professional in English and fluent in French. This includes writing, speaking, and reading

    Benefits 

    • Real Big Data experience with more than 70 million documents ingested every day, and a total of around 100 billion unique documents in storage 
    • Flat organization and strong culture 
    • Partial remote possible (up to 2 days per week). Gentilly-based office. 
    • International and diverse environment (US, EMEA, APAC) 
    • Staff canteen 
    • Complementary health insurance 
    • 10 RTT per year 
    • Synthesio allows employees to take time during their working hours for leisure (we had groups around Sport sessions together; Board games with the team; Free time ...). As long as work is done, you can organize your time as you want. 
    • Many team events are organized at the initiative of the team (Just to say people get along and there is a good ambiance) 

    Our Values 

    • Win As One Team: We are nothing without each other. We support each other, celebrate team spirit, and always move together. Be open minded, humble and a team player 
    • Ownership: Each team defines its own schedule of deliveries and methodology (from scrum to Kanban) and owns their projects from design to production. 
    • Test And Learn: We are not afraid to fail but we are afraid of not trying. We learn from our mistakes and always come back stronger 
    • Listen Up: The more we listen, the more we learn. Every person has something to teach up

Recruitment Process

Our Recruitment Process 

  1. A 30-minute screening phone call with the Team Lead CTO 

  2. A 60–90-minute interview with some engineers of the team. You will conduct a situational exercise (video call possible)

  3. Optional meet & greet (30 minute each) with various members of the team (Product Managers, Site Reliability Engineers, and Front-end engineers...) (video call possible) 

  4. Proposal

Additional Information

  • Contract Type: Full-Time
  • Location: Paris
  • Possible partial remote
Apply Now

See Other Ipsos Job Listings