We are almost at the end of a new year in Marvik. It was an excellent year, where the company grew a lot and we achieved many goals, but above all a year of personal and professional growth for the whole team, solving challenging technical problems and where we also had a lot of fun along the way. We want to share with everyone what this 2021 has left us and what we are expecting for next year.
We keep growing
Our team is what we are most proud of and what we care most about at Marvik. We care a lot about our culture, teamwork, and we really enjoy working together, solving problems with state-of-the-art technologies.
This year was a year of a lot of growth! We actually multiplied by 3 our headcount! We incorporated amazing people, while maintaining Marvik’s innovative spirit and team culture.
Another very important achievement this year is to be a more diverse team. Our company has people working from 6 different cities and more than fifth are women. One of our goals for next year, aligned with the gender equality goal from the Sustainable Development Goals (SDGs) of the United Nations, is to continue creating a more diverse team, increasing the number of women in it.
We are very grateful for the team we have at Marvik, they have done a great job this year, with a lot of commitment and professionalism.
After a long period of reduced traveling, finally in the last quarter of 2021 we were able to resume our travels!
We were in the US, visiting machine learning consulting clients and friends. Special thanks to all of you who welcomed us, it was great to be able to meet some of you in person, we had a great time!
We were also able to go to Argentina, to share with part of our team and we even went to the wedding of one of them! A very fun trip!
Also this year we had the opportunity to participate in the 500 Startups immersion program in San Francisco. We are very grateful for this amazing experience, we learned a lot with awesome mentors and were welcomed into an incredible community.
Clients, projects & tech
It was a year with very interesting, challenging and above all fun projects, technologies and clients. We have worked on almost 20 machine learning projects with different technologies to deliver state-of-the-art solutions. From forecasting to GANs and Transformers, we saw it all.
Just to name a few:
- Technologies like StyleGAN, Machine to machine translation, Lipsync, Pix2Pix and model distillation to generate realistic and efficient deepfakes (images and videos) for several clients.
- Recommender engines with millions of records using state-of-the-art tech like BERT Question Answering, embeddings and Elasticsearch.
- Edge computing projects with computer vision applications, using NVIDIA Jetson devices and Microsoft’s IoT Hub among others.
- Multiple Natural language processing projects using different tools and algorithms like Transformers, POS tagging, doc2vec embeddings, clustering, CLIP, BERTopic, Transformers Entailment, Camelot, GPT-3, Stanza. To name some of the projects:
- Understand and predict customer behavior using social media analysis
- Automatically extract key information from scientific documents of hundreds of pages
- Analyze and extract information from finance articles
- Structure e-commerce products catalog
- Also multiple interesting Computer vision projects like understand which products are similar to each other in e-commerce, automatically detect brands in different products, pose estimation, finding failures in production lines and others. For these projects we used CNNs, Image segmentation, Image Clustering, LIME interpretability, OpenCV, OCR, image embeddings.
- Several data science and data forecasting projects to predict stock shortages, optimize email campaigns, predict a patient’s next visit to the emergency room, predict prices and more.
Check out some of our work here.
Meetups & articles
During this year we continued researching, writing articles, organizing meetups and even helping as organizers at NeurIPS, one of the most important Machine Learning conferences.
The articles we published in 2021 included a variety of topics, from Computer Vision and GANs, MLOps, speech synthesis, electronics and ML, to what we need to consider before starting an ML project. Check some of them here.
- Speech synthesis and voice cloning
- How to optimize your model with TF-TRT for Jetson Nano
- Model distillation: leveraging lack of data with StyleGAN2
- MLOps: Just DevOps for machine learning?
- What you have to know, as a client, before starting a Machine Learning project
- When electronics meets machine learning
As ambassadors of deeplearning.ai, we continue to organize machine learning meetups open to the whole community.
Subscribe to our meetup group so you don’t miss the next events, or contact us if you want to share your machine learning experience or use case in an upcoming meetup!
BTW, thank you very much to everyone who subscribed and participated in the events, we are very proud to have created the largest ML community in our hometown on Meetup.
End of the year getaway
We had our end of the year getaway to Punta del Este with almost all the Marvik team! 🎉🎉
We shared a fun time together with barbecue, games and even swimming competitions in the pool. It was also the perfect excuse for some of our Argentine 🇦🇷 team members to come and meet us in person!
What about 2022?
Just as a preview, next year we plan to continue growing, always taking care of maintaining and strengthening the Marvik culture.
We offer state-of-the-art machine learning solutions for many of our clients, so keeping us always informed and up to date with new technological advances is a priority.
In addition, we want to be the company that best prepares engineers to work on machine learning projects. Our onboarding and training plan is already one of the best in the industry.