Senior Computer Vision Engineer/Scientist
Canada · India · Remote
Mid-level +1 · Full time
Posted 2 years ago
Apply

Help us tackle the growing wildfire crisis with the latest advancements in AI and IoT

Who we are

The problem: Every minute matters in fire response.  As climate change amplifies the intensity of wildfires—with longer fire seasons, dryer fuels, and faster winds—new ignitions spread faster and put more communities at risk. Today, most wildfires are detected by bystanders and reported via 911, meaning it can take hours to detect a fire, verify its exact location and size, and dispatch first responders. Fire authorities need a faster way to detect, confirm, and pinpoint fires, so that they can quickly respond—preventing small flare-ups from becoming devastating infernos.

About PanoPano is a venture-backed early stage climate tech startup that is the leader in wildfire early detection, leveraging the latest advancements in IoT, AI, satellites, and SaaS software to deliver actionable intelligence to customers. Pano leverages mountaintop cameras and satellites to detect the first traces of smoke and put real-time fire images in the hands of first responders to speed up containment. Pano is already partnering with major utilities, fire authorities, and government agencies in California, Colorado and Oregon, with plans to expand to other western states and Australia this year. Recent media coverage includes articles from the San Francisco Chronicle and TD World, a leading industry publication.

Our team is composed of seasoned technology professionals from companies such as Cisco, Apple, and Nest. Headquartered in San Francisco with an office and factory in the Mission District, our hybrid team works from locations around the world. Founded in mid-2020, we’ve raised over $8M from leading VC funds and prominent angel investors, including the CEOs of Gitlab, HomeLight, and People.AI. 

Role overview

Pano is looking for an experienced Computer Vision (CV) expert to develop and execute Pano’s roadmap for our smoke detection CV model(s) for DayTime Panoramic Camera, NightTime IR Camera and Satellite images.  He/she is responsible for developing CV models for early detection of Fire/Smoke in different types of cameras (Panoramic, IR and Satellite), maintaining database of fire/smoke labels, working with labellers and data engineering for implementing Active Learning pipelines, preprocessing and postprocessing to reduce the FP and FN and adopt the models for different geographic locations/landscapes and weather conditions. Ideal candidates have deep expertise in DL based computer vision, as well as experience with traditional computer vision techniques, time series, synthetic data, and near-real-time production inference. Position is remote but company meetings are scheduled for the Pacific time zone. This individual will report to Pano’s Engineering Manager of ML.

Key responsibilities

  • Design and develop CV models for smoke detection. Over time, these models will need to accommodate a growing number of camera feeds and camera types, as well as an ever-widening range of environmental conditions, while at the same time demonstrating continuous improvement in speed and accuracy.
  • Develop Time Series based CV models which take multiple frames as input in order to predict fire/smoke early. Design, train, evaluate and deploy Time Series Model and develop the active learning pipeline for continuous improvement over time.
  • Responsible for smoke detection models for production environment, including conducting experiments on new candidate models and assessing the performance of models in production as well as against a test set.  
  • Optimize the implementation of production near-real-time inference, working in partnership with a Python/full stack developer.
  • Oversee the training data cycle and management of training data, with support from multiple ML ops resources.
  • Contribute to intellectual property claims, research papers, and grant applications.
  • Track the state of the art in the field and investigate novel models, labeling techniques, and ML Ops tools that expand the scope and capabilities of our products.
  • Communicate the capabilities of our technology internally and externally in an engaging and educational way.
  • Support Pano’s commitments to data privacy and ethical AI.

Desired qualifications & experience

  • 2+ years recent experience in developing and implementing production deep learning models for computer vision tasks specifically object detection, including heavy involvement with or ownership of the infrastructure supporting model development (data, large scale training, eval, etc)
  • 3+ years of ML experience overall, with excellent ML fundamentals demonstrated by models successfully deployed in production
  • 1+ years of experience in Video object detection, classification, and tracking.
  • Experience working with traditional CV approaches, such as filters and optical flow.
  • Expertise in deep learning and associated frameworks, e.g. TensorFlow, PyTorch
  • Good SWE fundamentals and ability to demonstrate SWE experience in both high and low-level languages (Python and C++)
  • Bonus: Experience in deploying models in real time environments. Optimizing architectures to explore the accuracy vs. compute tradeoff 
  • Bonus: Experience working in a lean startup environment
  • Bonus: Masters or PhD in Computer Vision, Machine Learning, or other relevant field

Pano is an equal opportunity employer committed to recruiting and supporting our team-members regardless of where they come from. We do not discriminate on the basis of race, color, religion, gender, gender identity or expression, sexual orientation, national origin, genetics, disability, age, or veteran status.

Application Instructions

Interested candidates should forward a resume or a link to your LinkedIn profile to ML_Hiring@pano.ai.

Pano
Pano offers hardware and software technology solutions for fire professionals to detect threats and respond faster.
Size:  11-50 employees
© 2023 Pallet Labs Inc. See privacy policy and terms of service
Get started on Pallet
This community is on Pallet — Where creators turn their community into recruiting networks