Aeva is building the next-generation of sensing and perception for autonomous vehicles and beyond. With its unique ability to measure instantaneous velocity for each pixel, long range performance at high resolutions, while being free from LiDAR or sunlight interference, Aeva’s 4D LiDAR is built from the ground up at silicon photonics scale for mass market applications. Leveraging per point velocity and FMCW performance benefits Aeva’s customized Perception stack offers capabilities and performance metrics simply not available elsewhere in the market.
Aeva is seeking to hire a highly skilled and experienced Perception Verification and Validation Engineer to join our growing team! This individual will perform comprehensive characterization, analysis, verification, and validation of classical and machine learned perception software. Performance evaluations will cover a wide spectrum of Use Cases, environmental conditions and corner cases. Responsibilities will also include developing a very extensive database of real world and synthetic data collections, performing data curation and establishing ground truth data. The role involves directly interfacing with customers regarding performance issues and collaboration with Aeva’s SW / Perception development teams to address performance issues.
What you'll be doing:
- Verifying and Validating the performance of Aeva’s classical and ML Perception algorithms, including: detection, classification, segmentation, state estimation and localization.
- Work with the SW teams to enhance and augment post point cloud data products such as ego motion estimation, on-line calibration, object detection and classification, and scene segmentation.
- Characterize, verify and validate the performance of all Aeva data products - e.g. point clouds, intrinsic, extrinsic and on-line calibration, ego motion estimation and perception products.
- Develop, manage and execute processes and assets for data collection, data curation, ground truth determination and the analysis needed to verify and validate data products including localization and perception.
- Respond to the Customer’s performance concerns, including replicating, root causing and developing solution roadmaps.
What you have:
- BS with 2+ years working experience or a MS in characterization, testing, simulation, analysis, verification and validation of perception and state estimation SW.
- Deep knowledge and experience with systems engineering processes, including SW requirements, development and V&V.
- Strong background and experience in evaluating, testing and the analysis of classical and machine learned perception and/or state estimation algorithms.
- Experience developing automated test SW, with proficiency in Matlab and Python, as well as some familiarity with C and C++.
- Strong background in digital signal processing and machine vision algorithms including FFTs, Kalman Filters, clustering, classic machine vision and machine learning.
- Excellent written and verbal interpersonal skill with a strong attention to detail.
- Ability to prioritize and handle multiple projects at the same time with proven success in meeting deadlines.
Nice to have:
- Knowledge of photonics elements and their integration including: PICs - different material platforms, types of lasers, types of optical amplifiers, types of detectors, and free space optics.
- Experience and proficiency with C and C++.
- Knowledge and experience with automotive standards.
- Knowledge and experience with reliability and qualification testing
- Knowledge and experience with optical, silicon, electrical and mechanical manufacturing processes and testing from wafer level to module level FATP.
- Knowledge and experience developing machine learning algorithms.
What's in it for you:
- Be part of a start-up that's backed by industry leading venture capitalists
- Very competitive compensation and meaningful stock grants
- Exceptional benefits: Medical, Dental, Vision, and more
- Unlimited PTO: We care about results, not punching timecard
- Paid lunches, ping pong tournaments, and fun team off-sites