StartX is a business outsourcing company for today’s agile, technology-driven and fast-growing small and medium enterprises.
We own SAAS app that isdeveloped and managed by our in-house team.
The app processes a large number of documents to extract meaningful information into structured form and we are working to add new capabilities.
We are looking for an experienced computer vision engineer to join us in taking our application to the next level.
The selected candidate will be responsible for:
- Developing OCR, Machine Vision solutions to extract and overlay text and standardize the outputs (key value pairs).
- Implement image pre-processing algorithms to achieve the best OCR results from images.
- Transforming the output data appropriately to match the required data types.
- Designing and developing prototype software using the latest open-source technologies or proprietary SDKs.
- Analysing the efficiency of accuracy of different extraction tools and techniques to optimize the extraction accuracy and speed.
- Receiving and incorporating feedback from analysts, testers, and end users.
You should have a keen interest in building, deploying & supporting document processing services that extract and interpret information from invoices, emails, attachments and other sources.
A successful candidate will have:
- Bachelors in Computer Science
- At least three years’ experience working on OCR, Computer Vision and Machine Learning projects.
- Comfortable developing, reviewing and deploying Python and C# code
- Experience of working with leading cloud providers such as Azure Cognitive Services, Google Vision or AWS Textract.
- Experience of working with OpenCV or Tensorflow for computer vision
- Expert in handling the OCR output, storing and overlaying text overlay using JS Canvas or other suitable libraries.
- Structure the text output form OCR to develop meaningful outputs.
- Develop output standardization approach using Regix or other suitable templating techniques.
- Experience of working various OCR/Machine visions libraries, APIs and SKDs.