MODEL
EFFICIENCY
CONSTRAINT

YOUR MODEL. YOUR HARDWARE.

10× MORE EFFICIENT.

We convert your existing AI models to run on FPGAs at a fraction of the power — no retraining, no new hardware, no GPU dependency.

CURRENTLY ENGAGED WITH TIER 1 DEFENSE PRIMES AND INTERNATIONAL SYSTEMS INTEGRATORS
VIEW PERFORMANCEAPPLY
01 / Open Source

WHAT WE'VE SHIPPED

No. 01

SPIKE-SPACE-T1C

Unsupervised satellite detection from asynchronous event streams. Three-layer spiking convolutional network trained via STDP — no backpropagation required. Built for space domain awareness.

// EVENT-DRIVEN · STDP · INSTANCE SEGMENTATION

No. 02

SPECTRUM ANALYZER

Detects and classifies 11 radar signal types from spectrograms in real time. ResNet-18 backbone on modified YOLO architecture. 100% precision at confidence >0.9 on 512×512 inference.

// RADAR · RF CLASSIFICATION · REAL-TIME

No. 03

SPIKEYOLOV8-TRACKER

End-to-end pipeline for real-time object detection and tracking on event camera data. BICLab ECCV 2024 implementation with Hungarian-algorithm ByteTracker for temporal continuity.

// OBJECT DETECTION · BYTETRACKER · EVENT CAMERA

02 / Research

PUBLISHED WORK

OBJECT DETECTION

Event-Based Object Detection

5.7× EFFICIENCY GAIN

SpikeYOLO 5.7× efficiency improvement. Track fast-moving threats in real-time using 5× less power than conventional AI systems.

01
HUMAN-MACHINE INTERFACE

Gesture Recognition

75.76 GOP/s/W
02
AUTONOMOUS SYSTEMS

UAV Edge Control

7× OVER JETSON NANO
03
SPACE & RADIATION

Radiation-Tolerant Compute

5× HIGHER MTBF
04

// ADDITIONAL APPLICATIONS IN MEDICAL AND INDUSTRIAL AVAILABLE ON REQUEST

03 / Process

HOW WE WORK

03.1 INTAKE

BRING YOUR MODEL

Hand us your PyTorch or ONNX model as-is. No retraining. No pipeline changes. Your training workflow stays exactly the same.

03.2 DEPLOYMENT

10X MORE EFFICIENT

We handle conversion, optimization, and FPGA deployment. Your engineers interact with the output — not the conversion layer. No HDL expertise required on your end.

03.3 ONGOING

MANAGED SERVICE

We maintain the deployment after handoff. Optimization continues as your workload evolves. You own the outcome. We own the stack.

// ON THE ROADMAP: CUSTOM ASIC · 100× EFFICIENCY OVER JETSON · PIPELINE PARTNERS GET PRIORITY ACCESS
04 / Contact

WORK WITH US

DEFENSE

ACTIVE

SWaP-constrained platforms, autonomous systems, EW, radiation-tolerant compute

TELECOM

ACCEPTING PARTNERS

5G baseband, RF classification, sparse signal processing on FPGAs

INDUSTRIAL

ACCEPTING PARTNERS

Vibration, acoustic, and thermal sensor fusion at the edge

MEDICAL

RESEARCH STAGE

EEG, EMG, and implantable device inference — defining the problem together

Backed By

  • Israel Aerospace Industries
  • Stanford StartX
  • Antler VC
  • Mana Ventures
  • Ollin Ventures
  • Gaingels
  • Hulsey Richmond Ventures
  • NVIDIA Inception Program

Academic

  • Stanford University
  • SLAC National Accelerator Laboratory
  • University of Southern California
  • UC Santa Cruz
  • University of Milano-Bicocca
  • Istinye University

Contact

Apply

30-minute technical call

APPLY

© 2025 TYPE 1 COMPUTE. ALL RIGHTS RESERVED.

EDGE AI INFERENCE FOR PLATFORMS WHERE EVERY WATT COUNTS.