Mythic is building the future of AI computing with breakthrough analog technology that delivers 100× the performance of traditional digital systems at the same power and cost. This unlocks bigger, more capable models and faster, more responsive applications—whether in edge devices like drones, robotics, and sensors, or in cloud and data center environments. Our technology powers everything from large language models and CNNs to advanced signal processing, and is engineered to operate from –40 °C to +125 °C, making it ideal for industrial, automotive, aerospace, and defense. We’ve raised over $100M from world-class investors including Softbank, Threshold Ventures, Lux Capital, and DCVC, and secured multi-million-dollar customer contracts across multiple markets.
Mythic's analog compute hardware is a massive integration of analog and digital components on a single chip, with multiple cascaded digital and analog stages. The fundamental compute atomic is a vector-product that is entirely processed in the analog domain. So, analog impairments like ADC/DAC non-linearity and weight-noise directly impact the accuracy of the vector-multiply operation. Mitigation techniques for these impairments thus become critical and is the realm that the Systems Engineer operates in. Therefore, the Systems Engineer role maps directly onto skills from RF baseband, high-speed digital communication system design and RF sensing— think Wi-Fi, SerDes, gigabit Ethernet and sensor signal processing.
The Systems Engineering team
Sits at the intersection of Analog, AI, Firmware and Silicon Productization,
Models analog effects and their impact on neural network performance.
Develops signal-processing based solutions to mitigate impact of analog impairments on neural network accuracy
Works cross-functionally to validate, debug, and optimize analog compute hardware.
Contributes to the design of next-generation hardware.
Brings up new silicon, characterizes silicon performance and develops effective approaches for silicon screening
Builds frameworks for large-scale data capture and statistical error analysis for analog compute in the simulation domain and on actual silicon hardware
Here's what you will do
Own various aspects of algorithms and DSP blocks that optimize the performance of Mythic’s unique analog compute-in-memory technology from concept to customer deployment. This includes calibration loops, non-linearity compensation, offset-correction and estimation of residual-errors.
Work with model-training, compiler and firmware teams to productize these algorithms.
Write and modify firmware as needed to productize/debug algorithms
Continually improve on the fidelity of our modeling and simulation environment to better predict silicon performance.
Correlate errors seen on silicon to simulation models and contribute to improving the fidelity of our models for analog compute.
Develop Python frameworks for data collection, error-analysis and quantify impact of analog impairments on neural-network accuracy
Silicon bring-up, Characterization and Performance-Optimization.
Here's the background you need to have
Bachelor's degree in Electrical Engineering, Computer Engineering, Mathematics, Physics or a related field.
At least 5 years experience in production DSP or RF baseband engineering (< 3 years if Ph.D or M.S.)
Strong familiarity with production Python coding, including object oriented and/or functional programming
Strong familiarity with core DSP concepts, including frequency domain analysis, filtering, statistical signal processing and estimation theory
Track record of shipping silicon with DSP or RF/Analog sub-systems.
Understanding of linear algebra concepts, including matrix math and linear regression.
Comfort with large-scale collection and processing of signals.
Commitment to quality and engineering excellence.
Strong communication skills.
The following would be nice to have
MS/PhD in Electrical Engineering, Computer Science, Mathematics, Physics or related field.
Experience with RF calibration and silicon-bringup in the high-speed communication space
Strong familiarity with NumPy/SciPy (or experience with Numpy and strong familiarity with MATLAB for DSP).
Familiarity with state-of-the-art neural network architectures
