Computer Vision Engineer (Text Recognition)

Remote, USA

The Role

We’re looking for a brilliant Text Recognition and Classification specialist to help us become the next AdTech unicorn.

In this role you will be responsible for building core Computer Vision technologies in the areas of text recognition, media analysis and understanding. 

Experience in text detection and classification, ranking, and prediction are critical. Image enhancement and denoising, multiple-view geometry, text rotation detection, bundle adjustment, sensor fusion and other ML methods relevant for image/video analysis, object detection and recognition are also important.

To excel, you will have multiple years of industry experience in computer vision and machine learning (4+ years recommended) and 2+ years working with extracting and indexing text. Your Python, Linux, Keras, and AWS skills are peerless. You’re an expert in some combination of EAST, CRAFT, Tesseract, and Textract. You’re not intimidated by terabyte-scale datasets. And you are eager to assume an active leadership role at a fast-growing startup. 

Beyond technical chops, applicants must enjoy taking ownership while closely collaborating with others, enjoy iterating and pivoting quickly, and excel at building tools that are easy to understand and extensible.

Team & Technology

Deep.ad is a startup with a platform that detects, labels, and indexes marketing data (logos, products, pricing, etc.) in images and video. We bring established Big Data practices and scale to new-age media like TikTok and Twitch, empowering agencies to automate menial attribution work and generate meaningful insight.

We’re a fully-distributed team based in Chicago.

We adapt Scrum and Lean approaches depending on the project. Our architecture takes an API-first approach focused on independently maintainable microservices. We leverage different languages throughout our stack based on product constraints and goals.

We leverage different backend technologies including Python, Linux, Javascript, and Vue. Our infrastructure rests atop AWS, Azure and GCP. We leverage tools like Kubernetes, Cloud Functions, and Firebase. Our implementations are designed to be cloud-agnostic. All this activity is driven by data pulled from disparate sources. We use data to make our decisions, and we empower our clients to do the same. 

If you’re the type of person who comes to work every day expecting to learn, contribute, teach, take ownership and have fun then we think you’ll fit right in.

Responsibilities

  • Everything involved in applying a ML model to a production use case

  • Processing media at scale through computer vision and OCR

  • Designing and coding neural networks, gathering and refining data, training and tuning the model, deploying it at scale with high throughput and uptime

  • Take our models currently in production and identify areas for improvement; improve them using retraining and hyperparameter searches, then deploy without regressing on core model characteristics

  • Work with team to identify opportunities for improvement in our current platform and for enabling coming SaaS product launch

  • Ingest customer feedback and use that feedback to make frequent model improvements and develop new features

  • Actively contribute ideas for product improvements and solutions to technology challenges, including features and performance considerations.

  • Stay abreast of new ML technology and trends

Qualifications

  • 2+ years of experience with text recognition, classification, ranking, prediction, etc.

  • 4+ years of professional Machine Learning experience

  • 3+ years of experience with Linux, Tensorflow, and Python

  • You have successfully trained and deployed a deep learning machine model into production, with measurably improved performance over baseline

  • You are up-to-date on the latest deep neural net research and architectures, both in understanding the theory and motivations behind the techniques, as well as how to implement them in the ML framework of your choice

  • Experience prioritizing and performing multiple tasks in time-critical situations.

  • Comfort working within a fast-paced, dynamic and distributed environment.

  • Attention to detail, strong organizational skills and excellent follow-through.

  • Adept problem-solving ability, judgment and resourcefulness.

  • Strong written and verbal communication skills

  • Intellectual curiosity, self-motivation, independent with team building skills.

  • Graduate degree in Computer Science or similar technical field, with significant coursework in mathematics or statistics

Previous
Previous

Data Engineer, Web & Social Scraping

Next
Next

Machine Learning (ML) Engineer