Home 01 Technology 02 Projects 03 Research 04 About 05 GitHub ↗ Contact FuryLogic

Independent engineering company · AI · HPC · Systems

Engineering Intelligence. Building What Comes Next.

FuryLogic is an independent engineering company exploring the boundaries of artificial intelligence, high-performance computing, and intelligent software infrastructure.

Built and operated in-house Research with public DOIs Open source by default

FL/COMPUTE-LATTICE REV 2026.11 UNIT · 8 × NODES Δ 12.4 ms MEM TIER VECTOR KV SCHEDULER TRANSPORT
2.21× llama-server throughput at 64 concurrent users — Infernix
93.8% measured KV cache reduction at 65,536 tokens — OpusEdge
5.8× Redis SET throughput in durable benchmarks — StrikeDB
0.995 recall@10 over one million 768-dimensional vectors — StrikeDB

Measured in each project’s own benchmark suite. Sources are named on the project pages.

01 — The company

We build the layer below the product.

FuryLogic designs, writes and operates the infrastructure that intelligent software stands on — from CUDA kernels and storage engines to retrieval systems and the protocols between them.

FuryLogic is an independent technology and AI engineering company. Our work sits where performance is decided: inside inference engines, memory hierarchies, search pipelines and system runtimes. We are not a platform vendor, and we do not sell access to a model. We build software, publish the research behind it, and run it.

The portfolio is deliberately narrow in scope and deep in implementation. One team owns each system from the kernel to the HTTP surface, which means the person who wrote the scheduler is also the person who has to explain its tail latency. That constraint is the method.

We measure against named baselines on hardware people actually own — a single 8 GB GPU, a CPU-only server, a router with 4 MB of flash — because a result that requires a datacenter is not a result most operators can use.

“Engineering intelligence is not a slogan. It is the discipline of making a system explain itself.”

FuryLogic — engineering principles
What we are
An independent engineering and research company
What we build
AI systems, computing infrastructure, search, and systems software
How we publish
Open source where possible; preprints with DOIs for research
Where we operate
Self-hosted infrastructure, from kernel to interface

02 — Core disciplines

Six disciplines, one engineering practice.

The categories overlap on purpose. An inference engine is a systems problem; a search engine is a distributed-systems problem; both are judged by measurement.

01

Artificial Intelligence & Machine Learning

Model systems built to be measured, not merely demonstrated.

PyTorchONNXGGUF
02

LLM Inference Engineering

The work between a trained model and a served token.

RustC++20CUDA
03

High-Performance Computing

Memory hierarchies treated as the design constraint.

RustCUDAPCIe
04

Distributed Computing

Coordination that degrades predictably.

RustRESPPostgreSQL
05

Search & Retrieval Infrastructure

Retrieval you can evaluate and therefore improve.

PostgreSQLpgvectorBM25
06

Systems Programming

Kernels, runtimes and storage engines in Rust and C.

RustCMIPS assembly

Full technical focus

03 — Featured systems

Software that carries its own benchmarks.

Four systems in active development. Each one solves a problem we hit ourselves, and each one is measured against the tool it replaces.

All software — 11 projects

04 — Engineering philosophy

Correctness is a performance property.

We do not separate the engineering from the claim. A system that cannot be measured cannot be defended, and a benchmark that cannot be reproduced is a story.

Measure on real hardware

A number without a machine, a model and a baseline is a marketing claim. Benchmarks in this portfolio name the GPU, the CPU, the model file and the flags used to produce them.

Own the layer that matters

When an existing component cannot meet the requirement, we build the component. That is why this portfolio contains a database, an inference engine, a search stack and a memory manager rather than wrappers around them.

Make failure legible

Systems fail in ways their operators must understand. We prefer explicit refusals, typed outputs and honest gap audits over confident silence or a fallback that hides the problem.

Ship the whole thing

A kernel without a scheduler, an engine without a protocol and a paper without a reproducible path are incomplete. We carry work from the lowest layer to the interface a person actually uses.

Build the thing. Then prove it works.

Read about FuryLogic

05 — FuryLogic Open Research

Research published as research.

Papers are deposited with a DOI and stay citable. We link the publication, not a repository — the code that exists to support a paper belongs with the paper.

Paper 01 2026
DOI 10.5281/zenodo.21471506

OpusEdge: Telemetry-Guided Dynamic Compute Allocation for Dense, MoE, and Hybrid SSM-Attention Architectures

A single per-token importance signal, Δ, is extracted from dense hidden-state drift, native SSM selectivity or router-softmax entropy, then used to drive KV cache eviction, sparse attention, head gating and state compression without retraining. Measured at 93.8 % KV cache reduction at 65,536 tokens on one 8 GB GPU across dense, MoE and hybrid model families.

Irfan Mahir Zenodo preprint LLM inferenceKV cacheSparse attention
Paper 02 2026
DOI 10.5281/zenodo.21549808

VUGVA: A Software-Defined Virtual Unified GPU VRAM Architecture with CPU-Bypass Hybrid Memory for Non-NVLink Multi-GPU Clusters

GPU memory is fragmented across cards, and frameworks that spill through the CPU lose a large share of throughput. VUGVA virtualizes multi-GPU VRAM into one pool, transfers with 64-byte DMA descriptors so the CPU never touches payloads, and tiers memory from VRAM to DRAM to SSD with automatic promotion and demotion.

Irfan Mahir Zenodo preprint GPU virtualizationMemory tieringCPU bypass
Paper 03 2026
DOI 10.5281/zenodo.22810393

Attention Stream Cache: Reducing DRAM Bandwidth Saturation in LLM Attention Layers

Attention’s scattered KV cache reads carry an 80–84 % bandwidth penalty relative to sequential reads. ASC pairs a predictive attention router with an asynchronous streaming prefetch queue, staging the blocks a kernel will read next into L2 before it asks for them.

Irfan Mahir Zenodo preprint DRAM bandwidthPrefetchingCUDA

Four preprints on Zenodo · maintained by FuryLogic Labs

Enter Open Research

06 — Open source

The source is part of the argument.

Most of this portfolio is public: Rust, C, C++ and Python under MIT, Apache-2.0 and AGPL-3.0. Canonical repositories live on our own infrastructure, mirrored to GitHub.

  • Self-hosted Git at git.furylogic.com, with GitHub mirrors for reach.
  • Licenses chosen per project: permissive for infrastructure, copyleft where the implementation should stay open.
  • Papers carry their own license and a DOI, independent of the code.
  • Research prototypes are labelled as such. Nothing here is presented as production-ready when it is not.

Start a conversation

Building something that has to work?

We work on inference, infrastructure, retrieval and systems problems. If the constraint is performance, correctness or memory, that is the conversation.