Engineering Notes

The Neurmorph blog

Technical writing from the engineering team: compiler architecture, edge power budgets, SNN benchmarks, and the hard problems in neuromorphic computing software.

Architecture

Inside the NMC Compiler: Optimization Passes for Spike-Coded Graphs

Apr 1, 2026 · 16 min read

Company

Neurmorph 2025: What We Built and Where We're Going

Feb 10, 2026 · 6 min read

Edge Engineering

Event-Driven Sensor Fusion: Combining DVS Camera and IMU on a Sub-milliwatt Budget

Jan 6, 2026 · 10 min read

Architecture

From PyTorch to SNN: A Technical Conversion Walkthrough

Nov 25, 2025 · 15 min read

Edge Engineering

Why Autonomous Robotics Navigation Is the Killer App for Neuromorphic Inference

Oct 21, 2025 · 8 min read

Research

Temporal Locality as a First-Class Primitive in Event-Driven Attention

Sep 16, 2025 · 13 min read

Benchmarks

Our Benchmark Methodology: Measuring TOPS/W on Neuromorphic Silicon

Aug 12, 2025 · 14 min read

Edge Engineering

What Industrial IoT Teams Actually Need from Edge AI Inference

Jul 8, 2025 · 7 min read

Architecture

Designing a Hardware Abstraction Layer for Heterogeneous Neuromorphic Chips

Jun 3, 2025 · 11 min read

Research

SNN vs ANN Latency Trade-offs at the Edge: A Practitioner's Guide

Apr 29, 2025 · 10 min read

Edge Engineering

The Physics of Coin-Cell Inference: Energy Budgets at the Microwatt Frontier

Mar 25, 2025 · 9 min read

Architecture

Building a Neuromorphic Model Compiler from a PyTorch Frontend

Feb 18, 2025 · 12 min read

Architecture

Why Spike Coding Beats 8-bit Quantization for Always-On Inference

Jan 14, 2025 · 8 min read