gli.al

Infrastructure consulting and operations

What we do

gli.al plans, builds and runs infrastructure for teams that cannot afford to get it wrong. We bring the discipline of validator operations, where an outage or a mishandled key costs money in public, to ordinary servers, networks and services, centralized or decentralized.

  • Consulting & architecture

    Reviews of an existing setup, architecture and migration plans, or a second opinion before you commit.

  • Hosting & operations

    Bare metal and cloud, declared in NixOS, patched, backed up and restorable.

  • DNS & network design

    Zones and DNSSEC, addressing plans, IPv6, WireGuard and firewalls.

  • Monitoring & alerting

    Metrics, logs and alerts that mean something, with escalation and response times agreed per engagement.

  • Key & secrets management

    Hardware-backed keys, from the key ceremony to signing policies and rotation.

  • Security & hardening

    Hardened kernels, sandboxing, access reviews and incident response.

How we run it

Whatever we take on is run the same way, whether we keep operating it for you or hand it over.

  • Safe & secure

    Best practice and defense in depth: hardened kernels, sandboxing and access keys stored in hardware.

  • Monitored

    Nothing goes live without logs, metrics and alerts, so problems surface before your users report them.

  • Redundant

    Spread across providers and data centers, with no single point of failure where it matters.

  • Reproducible

    Reproducible, declarative and reliable deployments thanks to NixOS.

  • No lock-in

    Declared in code and documented, so your team or another operator can take over at any time.

  • Free as in speech

    We build on free software and publish our improvements back to the community.

Where it comes from

We learned this running validators on Polkadot and Kusama: nodes that sign with keys worth protecting, are measured in public block by block, and are penalized for every outage and every double-sign. The validators are retired; the habits are not. Keys live in hardware, nothing depends on a single machine or provider, alerts fire before users notice, and every deployment can be rebuilt from its description.

gli.al is the infrastructure practice of F.R.E.D., a studio in Zug, Switzerland, with a background in leading engineering and infrastructure for Substrate-, Ethereum- and Solana-based networks through to mainnet launch. The same care goes into a mail server, a build pipeline or a company network.

Our name

We named ourselves after neuroglial cells, which are responsible for supporting and protecting neural cells across the nervous system. Without their support, the billion-node decentralized network of neurons that drives all information processing in the central and peripheral nervous system could not exist. We aim to be that for the systems you depend on.

Get in touch

Most work starts as a review of what you have, continues as a build we hand over documented, or stays with us as managed operations. Tell us where you are.