The posting
TITLE: Computer Vision Engineer
LOCATION: Gurugram/Bangalore WFO.
Build the intelligence behind real-world video.
At Awiros, we’re building an operating system for Computer Vision - a platform that enables developers and enterprises to build, deploy, and scale AI-powered video applications without having to reinvent the infrastructure underneath.
Founded in 2015 and backed by Series A funding (~$7M raised), we work at the intersection of Deep Learning, Computer Vision, distributed systems, and real-time video analytics. Our technology powers applications across detection, recognition, tracking, safety, security, and operational intelligence.
And this isn’t Computer Vision in a notebook.
Our models run on real-world video, at scale, across edge and cloud environments — where accuracy, latency, robustness, and compute efficiency all matter.
That’s where you come in.
THE ROLE:
We’re looking for Computer Vision Engineers who enjoy turning challenging perception problems into reliable, production-ready solutions. You’ll work across the Computer Vision and Deep Learning stack - building and fine-tuning models, evaluating performance, optimizing inference, & integrating solutions into real-world video analytics applications.
The challenge isn’t simply getting a model to work. It’s making it work reliably in the real world — across different environments, hardware configurations, and deployment conditions, while balancing accuracy, latency, memory, throughput, and reliability.
You’ll work closely with our Computer Vision, Platform, Product, and C++ engineering teams to take solutions from implementation through optimization and into production.
WHAT YOU'LL WORK ON:
- Design, build, and optimize vision algorithms and deep learning models for real-time video analytics.
- Develop solutions across object detection, tracking, re-identification, pose estimation, recognition, segmentation, and anomaly detection.
- Select, adapt, and fine-tune deep learning architectures based on the requirements of a specific problem and deployment environment.
- Work with real-world datasets to train, evaluate, benchmark, and improve model performance.
- Analyze model and inference bottlenecks across accuracy, latency, memory, throughput, and compute utilization.
- Optimize models for deployment using techniques such as quantization, pruning, model simplification, and architecture optimization.
- Collaborate with C++ and systems engineers to integrate models into production pipelines and deliver validated model artifacts such as ONNX models.
- Debug challenging detection and perception failures, identify root causes, and build robust solutions.
- Develop evaluation workflows and performance benchmarks to measure and continuously improve our Computer Vision systems.
- Evaluate new Computer Vision techniques and technologies based on their practical applicability and potential product impact.
- Take ownership of problems from implementation → validation → optimization → integration → deployment.
WHAT WE'RE LOOKING FOR
You don’t need to know everything. But you should have strong technical fundamentals, practical implementation skills, and enjoy figuring things out when the answer isn’t obvious.
Core Skills
- Strong fundamentals in Machine Learning, Deep Learning, and Computer Vision.
- Good understanding of the mathematics behind ML - particularly linear algebra, probability, and calculus.
- Hands-on experience with PyTorch or TensorFlow.
- Solid understanding of Computer Vision concepts including image processing, feature extraction, object detection, tracking, segmentation, and recognition.
- Strong programming skills in Python.
- Practical experience with OpenCV and NumPy.
- Familiarity with scikit-learn is a plus.
It’s a Plus If You Have
- Experience with C++ and/or GPU programming.
- Experience deploying models using TensorRT, ONNX Runtime, or similar inference frameworks.
- Exposure to model optimization and edge AI.
- Experience working with data pipelines, annotation workflows, evaluation frameworks, or large-scale datasets.
- Familiarity with modern architectures such as Transformers, VLMs, or multimodal models.
- Experience taking models from prototype to production.
- Experience diagnosing performance issues across the model, inference stack, and underlying hardware.
- Contributions to open-source ML/CV projects, technical competitions, or strong personal projects demonstrating practical implementation skills.
WHY AWIROS?
Work on Computer Vision that leaves the lab.
- Build models that operate on real cameras, real environments, and real-world constraints - not just curated datasets.
Go deep technically.
- Work alongside engineers across Deep Learning, Computer Vision, C++, GPU computing, inference optimization, distributed systems, and video infrastructure.
Own the entire journey.
- Get exposure to the complete model lifecycle - from implementation and experimentation to optimization, deployment, and production performance.
Solve problems that aren’t already solved.
- Video intelligence throws up messy, open-ended problems. There’s room here to challenge assumptions, experiment with different approaches, and build solutions that work reliably in the real world.
Build what comes next.
- Computer Vision is moving quickly. At Awiros, you’ll evaluate and adopt emerging technologies where they create real value - and turn them into practical, scalable video intelligence applications.
If you’re excited by the idea of making machines understand the visual world - and want to work on the engineering required to make that happen at scale - we’d like to hear from you!
