Senior Embedded Engineer, DSP Algorithms at Gridware in San Francisco, CA | Kablio

Role

Description

profiling

debugging

model refinement

embedded code

firmware integration

monitoring

Profile runtime behavior, identify bottlenecks, and perform low-level debugging.
Collaborate with algorithm / ML researchers to refine models for embedded deployment.
Convert DSP algorithms and build ML inference pipelines into efficient embeddedC/C++ code for microcontrollers or other constrained platforms.
Work with firmware teams to integrate algorithms into system software.
Develop monitoring and observability systems to track model performance, data drift, data quality, and overall system health.
Optimize code for memory footprint, CPU usage, and real-time performance.

Requirements

c/c++

arm cortex-m

cmsis-nn

tensorflow

bs/ms

dsp

We are looking for a highly skilled Embedded Engineer who can translate advanced DSP algorithms and machine learning models into efficient, production-ready C/C++ implementations optimized for extremely resource-constrained environments. You will work closely with ML scientists and firmware teams to bring cutting-edge signal processing capabilities onto embedded platforms with strict memory, computing and power budgets.

Ability to read/translate algorithmic descriptions in Python/Matlab into low-level code.
Experience with ARM Cortex-M or similar MCUs.
Knowledge of low-level optimization techniques such as pipeline-aware coding, lookup table design, and memory layout optimization.
Experience working in extreme resource-restricted systems.
BS/MS in Electrical Engineering, Computer Engineering, Computer Science, or related field.
Experience translating and optimizing machine learning models for embedded targets (e.g., quantization, fixed-point, pruning).
Experience pushing on-device ML models to production (C++)
Familiarity with microcontrollers, RTOS, SoCs, or custom hardware.
Strong proficiency in C/C++ for embedded systems.
Hands-on experience with on-device ML frameworks (CMSIS-NN, etc.).
Understanding basic DSP concepts (filters, FFTs, spectral processing, etc.)
Experience in common ML frames (TensorFlow, PyTorch, Boosted Training, etc.)

Benefits

Company-paid training

“Off the Grid”, a two week per year paid break for all employees.

Health, Dental & Vision (Gold and Platinum with some providers plans fully covered)

Alternating day off (every other Monday)

Commuter allowance

Paid parental leave

Training + Development

Information not given or found

Interview process

Information not given or found

Visa Sponsorship

Information not given or found

Security clearance

Information not given or found

Company

Overview

$26.4M Series A

Funding Round

Secured significant investment from notable firms including Sequoia, Lowercarbon, and Fifty Years.

70% Outage Reduction

Service Improvement

Customers experienced a dramatic decrease in the duration of power outages thanks to Gridware's technology.

10,000+ Poles

Deployment Scale

The system has been installed across a vast network spanning thousands of poles and miles of power lines.

7× Revenue Growth

Financial Achievement

The company experienced rapid expansion and success in its first year of operation.

Founded by a former lineman and two Berkeley engineers to revolutionize power grid awareness.
Flagship Gridscope device provides edge analytics and alerts even during outages.
Technology integrates multiple types of sensing to detect hazards like fallen limbs.
After a pilot with PG&E, expanded operations across North America.
Physical tests include extreme conditions like live wire cutting to optimize hazard detection.
Combines solar-powered, ruggedized sensors with a cloud-independent analytics system.

Culture + Values

Passion, innovation, and technology align to modernize the power grid.
Pioneering ‘Active Grid Response’—immediate detection, precise localization, detailed identification, outage monitoring.
Developed in partnership with utilities—focused on working alongside linemen.
Built by a lineman for linemen—understands the rigor and commitment required in the field.
Relentlessly committed to giving crews the right tools to get the job done.

Environment + Sustainability

1,000+ miles & 10K+ poles

Gridscope Deployment Scale

Expanded Gridscope sensor deployment covers over 1,000 miles of power lines and 10,000 poles, enhancing wildfire risk mitigation efforts across multiple states.

70% Outage Reduction

Improved Grid Reliability

Gridscope technology has been shown to significantly reduce outage durations by over 70%, enhancing power grid stability and wildfire prevention.

Technology deployed across five U.S. states to reduce wildfire risk via early detection of vegetation and fallen-line hazards.
Continuous environmental stressor measurement (e.g., vibrations from falling branches) enables predictive maintenance and environmental hazard mitigation.

Inclusion & Diversity

Articles

About Us

Careers

Contact

Role

Description

profiling

debugging

model refinement

embedded code

firmware integration

monitoring

Profile runtime behavior, identify bottlenecks, and perform low-level debugging.
Collaborate with algorithm / ML researchers to refine models for embedded deployment.
Convert DSP algorithms and build ML inference pipelines into efficient embeddedC/C++ code for microcontrollers or other constrained platforms.
Work with firmware teams to integrate algorithms into system software.
Develop monitoring and observability systems to track model performance, data drift, data quality, and overall system health.
Optimize code for memory footprint, CPU usage, and real-time performance.

Requirements

c/c++

arm cortex-m

cmsis-nn

tensorflow

bs/ms

dsp

Ability to read/translate algorithmic descriptions in Python/Matlab into low-level code.
Experience with ARM Cortex-M or similar MCUs.
Knowledge of low-level optimization techniques such as pipeline-aware coding, lookup table design, and memory layout optimization.
Experience working in extreme resource-restricted systems.
BS/MS in Electrical Engineering, Computer Engineering, Computer Science, or related field.
Experience translating and optimizing machine learning models for embedded targets (e.g., quantization, fixed-point, pruning).
Experience pushing on-device ML models to production (C++)
Familiarity with microcontrollers, RTOS, SoCs, or custom hardware.
Strong proficiency in C/C++ for embedded systems.
Hands-on experience with on-device ML frameworks (CMSIS-NN, etc.).
Understanding basic DSP concepts (filters, FFTs, spectral processing, etc.)
Experience in common ML frames (TensorFlow, PyTorch, Boosted Training, etc.)

Benefits

Company-paid training

“Off the Grid”, a two week per year paid break for all employees.

Health, Dental & Vision (Gold and Platinum with some providers plans fully covered)

Alternating day off (every other Monday)

Commuter allowance

Paid parental leave

Training + Development

Information not given or found

Interview process

Information not given or found

Visa Sponsorship

Information not given or found

Security clearance

Information not given or found

Company

Overview

$26.4M Series A

Funding Round

Secured significant investment from notable firms including Sequoia, Lowercarbon, and Fifty Years.

70% Outage Reduction

Service Improvement

Customers experienced a dramatic decrease in the duration of power outages thanks to Gridware's technology.

10,000+ Poles

Deployment Scale

The system has been installed across a vast network spanning thousands of poles and miles of power lines.

7× Revenue Growth

Financial Achievement

The company experienced rapid expansion and success in its first year of operation.

Founded by a former lineman and two Berkeley engineers to revolutionize power grid awareness.
Flagship Gridscope device provides edge analytics and alerts even during outages.
Technology integrates multiple types of sensing to detect hazards like fallen limbs.
After a pilot with PG&E, expanded operations across North America.
Physical tests include extreme conditions like live wire cutting to optimize hazard detection.
Combines solar-powered, ruggedized sensors with a cloud-independent analytics system.

Culture + Values

Passion, innovation, and technology align to modernize the power grid.
Pioneering ‘Active Grid Response’—immediate detection, precise localization, detailed identification, outage monitoring.
Developed in partnership with utilities—focused on working alongside linemen.
Built by a lineman for linemen—understands the rigor and commitment required in the field.
Relentlessly committed to giving crews the right tools to get the job done.

Environment + Sustainability

1,000+ miles & 10K+ poles

Gridscope Deployment Scale

Expanded Gridscope sensor deployment covers over 1,000 miles of power lines and 10,000 poles, enhancing wildfire risk mitigation efforts across multiple states.

70% Outage Reduction

Improved Grid Reliability

Gridscope technology has been shown to significantly reduce outage durations by over 70%, enhancing power grid stability and wildfire prevention.

Technology deployed across five U.S. states to reduce wildfire risk via early detection of vegetation and fallen-line hazards.
Continuous environmental stressor measurement (e.g., vibrations from falling branches) enables predictive maintenance and environmental hazard mitigation.

Inclusion & Diversity

Articles

About Us

Careers

Contact