Sr S/W Engineer Machine Learning P-077

 SMASH, Who we are?

We are agents for tech professionals in Costa Rica and Colombia that help them build careers in the United States. 

We believe in long-lasting relationships with our talent. We invest time getting to know them as individuals and understanding what they are looking for as their professional next step. 

We aim to find the perfect match. As agents, we make sure to pair our talent with our US clients, not only by their technical skills but as a cultural fit. Our core competency is to find the right talent, fast.

We purposefully move away from the “contractor” or “outsourcing” type of relationship. Our clients don’t want contractors or “just a service.” Neither does our talent.

Our Benefits

• Work from everywhere
• Private Medical Insurance
• English Academy for Employees and Relative
• Business Skills Coach – Certifications
• Discounts with Tech Universities
• Events and additional Perks

This is a remote position for Costa Rica and Colombia
This position is Remote to work with a US Company, you will require to have Citizenship or work permit from Costa Rica or Colombia to apply for this role

Responsibilities

• Build training/labeling tooling and infrastructure
• Support data labeling/training/deployment efforts
• Develop domain-adapted model architectures/hyperparameters
• Explore new ML approaches to business problems/opportunities 

Requirements:

• 3+ years of experience (either in academia or industry)
• Bachelor's degree or above in computer science, mathematics, or a related discipline
• Experience with frameworks like Pytorch or Tensorflow
• You have taken the initiative to improve some or all of a deep learning pipeline in the past: dataset construction & curation, model design & iteration, and/or validation design & implementation. You have rethought approaches, earned approval, and taken action.
• You enjoy building tooling people actually use and like to help people use those tools with helpful documentation and other assistance.
• You value humility and seek out course-correcting information, whether that be unexpected learning behavior detected with experiments/visualizations, or domain-specific considerations that the experts we work with know.
• You seek out promising new developments in ML but prudently judge long-term cost/benefit tradeoffs before implementing sweeping changes.