Tesla

Makes electric cars, robotaxis and grid batteries

Computer Vision & Machine Learning Engineer

Computer VisionUpdated on Nov 11

Full-timePalo Alto, CA, USANo salary listedIn-person

About the Job

Requirements

Several years of experience shipping computer vision into a real operation, beyond notebooks and benchmarks.
Factory, automated optical inspection, electronics, or other industrial inspection experience.
Experience in PCBA and electronic assembly manufacturing, including IPC workmanship and inspection standards.
Hands-on experience in detection, segmentation, classification, or visual anomaly detection, including labeling, class imbalance, domain shift, and threshold setting.
Ability to evaluate a model using escape rate, false rejects, and cycle time, not only accuracy or mean average precision.
Judgment to recommend a commercial system when it is the better path and to distinguish a vendor demonstration from a result on the employer's images.
Ability to work with design, process, and quality engineers and turn their defect definitions into a testable problem.
Proficiency in Python and a standard training stack.
Experience deploying a model to a station, edge device, or factory server.
Evidence of stopping an approach because the images or the operation could not support it.

Responsibilities

Assess line images for camera, lighting, resolution, and consistency across lines and products, and determine whether defects are visible and separable from normal variation.
Turn the quality problem into a testable vision problem, including which defect is in scope, the acceptable miss rate and false-reject rate, and the required cycle time.
Build the minimum dataset and labeling approach for rare defects, weak labels, and a large volume of good images.
Select classification, detection, segmentation, or anomaly detection based on the data rather than a predetermined architecture.
Run a time-boxed proof on the employer's images, comparing real options including classical vision, a trained model, and a commercial automated optical inspection or vision system.
Recommend whether to build or buy, including what data, lighting, or camera changes are required before scaling.
Pilot the chosen approach on one line, covering latency, station integration, traceability, and monitoring.
Validate the chosen system on production data, covering holdout images, line trials, golden-sample checks, and other vision-system validation methods as required.
Define the retraining path for product or process changes.
Report negative results with evidence when the images or the line cannot support the decision.
Identify a solution in collaboration with supplier industrialization engineers, internal stakeholders, and external solution providers; the role does not receive a labeled dataset or a predetermined architecture.

Category	Supply Chain
Location	PALO ALTO, California
Requisition ID	288552
Type	Full-time

Job Description

<p>Own the path from line images to a solution Tesla can trust. We collect images on PCBA and TLA lines and developing models that to do this. Your job is to determine what the images can support, prove or kill an approach on our data, and leave us with a solution that works in production, whether that is a model we run, a system we buy, or a clear stop. </p><p>You will work with Supplier industrialization engineers, and internal Tesla stakeholders, external solution providers. You will be expected to identify a solution; you will not be handed a labeled dataset or a chosen architecture. </p>

Job Responsibilities

Assess line images for camera, lighting, resolution, and consistency across lines and products, and determine whether defects are visible and separable from normal variation

Turn the quality problem into a testable vision problem, including which defect is in scope, the acceptable miss rate and false-reject rate, and the cycle time the result must meet

Build the minimum dataset and labeling approach for rare defects, weak labels, and a large volume of good images

Select classification, detection, segmentation, or anomaly detection based on the data, not a predetermined architecture

Run a time-boxed proof on our images, comparing a small set of real options, including classical vision, a trained model, and a commercial AOI or vision system

Recommend build versus buy, including what data, lighting, or camera changes are required before scaling

Pilot the chosen approach on one line, covering latency, station integration, traceability, and monitoring

Validate the chosen system on production data, covering holdout images, line trials, golden-sample checks, and other vision-system validation methods as the case requires

Define the retraining path for product or process changes

Report negative results with evidence when the images or the line cannot support the decision

Job Requirements

Several years shipping computer vision into a real operation, not only notebooks and benchmarks

Factory, AOI, electronics, or other industrial inspection experience

Experience in PCBA and electronic assembly manufacturing, including IPC workmanship and inspection standards

Hands-on experience in detection, segmentation, classification, or visual anomaly detection, including labeling, class imbalance, domain shift, and threshold setting

Ability to evaluate a model on escape rate, false rejects, and cycle time, not only accuracy or mAP

Judgment to recommend a commercial system when it is the better path, and to distinguish a vendor demo from a result on our images

Ability to work with design engineer, process and quality engineers and turn their defect definitions into a testable problem

Proficiency in Python and a standard training stack

Experience deploying a model to a station, edge device, or factory server

Evidence of stopping an approach because the images or the operation could not support it

About the Company

Tesla

Makes electric cars, robotaxis and grid batteries

Tesla makes electric cars and batteries that store power for homes and utilities, and it develops self-driving software and humanoid robots. In 2023, its Model Y became the first electric vehicle to rank as the world's best-selling car. Buyers order the Model 3 sedan, Model Y SUV, or Cybertruck pickup directly from Tesla instead of through dealers, and they charge on its Supercharger fast-charging network. It also sells Megapack and Powerwall batteries, runs a robotaxi service, and charges drivers a monthly fee for self-driving software that still needs their supervision. Tesla was founded in 2003, with Elon Musk as its lead early investor and a recognized co-founder, and it is based in Austin, Texas.

Company size: 10,001+
Company stage: IPO
Headquarters: Austin, TX
Founded: 2003

Recently Posted Jobs

Explore more opportunities at Tesla

See more jobs