Applied Computer Vision Engineer

As an Applied Computer Vision Engineer, you will be working on the state-of-the art challenges in Computer Vision & solving them with novel algorithms and optimizations. The majority of our work is focused on applied Computer Vision - so a good understanding of the current state of the art of models (Classification, Object detection, Object Recognition, OCR, LayoutML, GAN etc networks) is mandatory.

We’re looking for someone who loves to solve problems and enjoys building algorithms and models that identify all possible scenarios. You must enjoy breaking down a problem statement into smaller chunks and have a knack of designing scalable systems using modular pieces.

A day in your role will involve any of the following (or a combination):

• Understanding business objectives and developing Computer Vision based solutions that help achieve it. The software could involve training framework, inference framework, working with different technologies for ML.
  

• Understand the requirements in depth and give adequate time estimates for the features
  

• Develop processes for different common operations of the team: data acquisition, model training, prototype development.
  

• Finding open-source datasets for prototype development.
  

• Develop pipelines for data processing, augmentation, training, inference and active retraining.
  

• Training models and tuning their hyperparameters
  

• Analyzing the errors of the model and designing strategies to overcome them
  

• Deploying models to production
  

• Optimize models for performance at the edge
  

• Coordinate with QA for every feature and make sure releases are bug free
  

This list is not exhaustive at all, and you'll have a lot of autonomy over your work.

This is a full time position based out of Noida. During the interview, you'll be able to share what you're most interested in.

Why you might be excited about us

1. We're working to solve a massive global problem and help organizations be more competitive. We love solving problems using technology and are bridging a massive gap in the operations excellence domain.
   

2. We're small, so you'll be able to contribute efficiently and without bureaucracy. You'll quickly have responsibility over big areas of our product.
   

3. Our team has a wide range of experiences (Oilfield, Telecom, Fintech, etc.) and are excited to learn from you, too.
   

4. You'll be able to work independently and set your own schedule. We don't micromanage and will help you do great work. We trust our people and believe that each person here puts their best foot forward.
   

5. We'll mold your role to shape a career you're excited about. We care a ton about your satisfaction and job happiness, and will help prep you for whatever you're looking for in the future.
   

6. We work hard and also highly value balanced work/life. We care about family and your own personal development, and don't expect for you to be always engaged with work.
   

Why you might not be excited about us

1. We're small (40+ people in the company now), so if you like more established companies, it's not (yet) the right time. You'll help build our company's culture.
   

2. Since we're an early stage startup, projects and priorities may shift.
   

3. Our customers love us, and there's a lot we can improve. It's a great place to be, but it means there's some jank. (Nothing too scary! ?)
   

4. Since you'll have a lot of responsibility and creativity over projects, they may not be defined perfectly initially. You'll be expected to bring your own experience and perspective to help us do the right things, and raise flags if you think we should do things differently.
   

About you

None of these are requirements, but do describe the kinds of people that we think would be most effective at Syook right now.

1. Love thinking broadly about problems and thinking creatively about how to solve them efficiently.
   

2. Happy to try things out to validate new features, and move on if they no longer solve a problem.
   

3. Excited for a front-row seat into a fast growing, early stage company. Things will change a lot!
   

4. Enjoy thinking through trade-offs, with both mindfulness of short-term needs and our long-term direction.
   

5. Happy writing documentation so that others can ramp up super easily and you're never a single-source-of-failure. We're a bit too small to have silos.
   

6. You are driven and care about doing a good job and improving your craft.
   

7. You have a growth mindset, can keep up with the latest technology changes and trends and suggest enhancements based on these
   

8. Most importantly, you’re the kind of person who is friendly, approachable, ready to help others and personally driven to put your best foot forward.
   

Technically these are some of the things we would like for you to have to be able to contribute effectively in this kind of a position:

1. Bachelors/Masters degree in Computer Science/Computer Engineering/IT or related fields
   

2. 1-2 years of experience. Exceptional candidates with less experience are welcome to apply
   

3. Sound knowledge of various deep learning classification models and object detection models
   

4. Experience of working with Image & Video Processing (OpenCV, Scikit, GStreamer, Tensorflow, PyTorch, NVIDIA Deepstream, TensorRT, Keras, Model Training/Inference, Video Processing Pipelines, Different GStreamer Convertors etc)
   

5. Good knowledge of statistics relevant to Deep Learning and ability to use the right statistical measures to assess model’s performance
   

6. Strong experience with Python and writing reusable code
   

7. Experience with Python web framework, e.g. Flask, Django or FastAPI
   

8. Experience with databases (MongoDB, Postgres, etc)
   

9. Experience with cloud environments for software development and deployment (AWS, Azure, GCP)
   

10. Experience with utilizing various GPU-based training infrastructures
    

11. Experience with Docker
    

12. Knowledge of DevOps and MLOps best practices for production Machine Learning systems
    

13. Experience in optimizing and pruning models, converting models to NVIDIA TensorRT etc would be preferable
    

14. Experience with Azure, Azure ML Studio framework etc. will be preferred
    

15. Prior experience in taking at least one solution to production from Concept to Deployment
    

16. Fluent in using the tools of the trade: Testing, Editors, Git, Command line, Slack
    

Our current development practices

Since we're an early-stage startup, we constantly have to ask "what gets the most value, cheaply, to validate our assumptions?" We build some things to last a long time, and others as prototypes.

Our development practices involve product specing and designing, agile sprint based development, e2e testing, CI CD, observability logging, and production probers. We've documented our product and code repos so that you should be able to get started easily—and if there are improvements needed, we'll absolutely improve our docs.

We recognize the value of maintainability and keeping our developer experience nimble. Our sprints are for 2 weeks and we push releases to production as per this schedule.

You'll help push us to be our best, and we're excited for recommendations and insights you have as you join. You'll be an owner and contribute towards how we work.

Interview process

We want you at your best, and won't be giving you gotcha-style algorithms questions. We want to get to know you, hear about what you're interested in, and learn about what you hope to do in the future.

1. Meet us and learn about Syook

You'll first talk to someone from the People Services team (over phone or video) and won't need to prepare anything in advance. The goal of this conversation is to get to know you and mutually explore if we might be a good fit for each other. You'll learn more about Syook and have a chance to ask any questions about our company, team culture, and product.

We'll also talk in depth about your technical experience. We'll talk about frameworks you've used, how you make technical decisions, types of problems you like to solve, etc. You won't need to prepare anything in advance.

This is primarily used to get a better feel for your experience, how you work, and where you may fit in. It'll be used to design the rest of the interview process.

Through this and the next steps, you'll meet more people in the company so we can get to know each other.

2. Technical Discussions
There will be a few rounds of technical discussions where we interact with you to understand in depth about your craft. These discussions will be with our CTO (Aman Agarwal). We will try and see how you can augment our engineering culture and we’ll discuss the bigger picture also. You will also get to understand what your role is all about and how you can grow with us. Please bring forth all previous work that you have done to these discussions to help us better understand what you have built and achieved.

3. People & Culture fit round
If you reach this stage, it automatically means that we’re convinced of your technical skills. However, that’s never enough for us. We want to ensure that you will feel comfortable working with us and that we can give you an environment where you can be your productive best. We will use this discussion to understand what you bring to the table apart from your technical skills in terms of initiatives, personality, and a certain entrepreneurial mindset.

4. Reference conversations
We will talk to a few people you've worked with before to learn more about how we can best work with you. We expect to hear great things, so this is primarily so we can work with you as effectively as possible.

Getting Started

If you're excited to learn more, you can directly mail us ([email protected]). If we think you might be a fit, we'll respond really fast. Please let us know if you have timing constraints. Regardless, we'll try our best to respect your time along the process.

You can learn more about what we're up to:

1. Our company’s website () to learn more about the space we're working in, where we write about how IoT is a gamechanger to build smart factories and workspaces.
   

2. Check out our social media handles like Facebook, LinkedIn, Twitter and Instagram (