Skip to content
Last Furlong
Language: English

Private AI integration

We handle the last mile of AI.

We build AI agents that answer questions against your own systems — your databases, APIs and file shares — running where your policy allows: on your hardware, on an approved hosted model, or fully hosted by us. The model was never the hard part. Wiring it in and getting it past your security review is.

Model choice, prompt design, benchmark scores. The part that demos well.

Deployment, integration, security review. The part that ships.

Two and a half minutes of it working

An agent is asked a question in plain English, discovers a database schema it was never given, writes the query itself and returns the answer — then we check what left the network. Every command and every line of output is real.

Served from this domain, not an embed — no third-party player, no tracking.

AI pilots rarely fail on model quality.

They stall at the boundary. By the time a pilot is ready to become something people depend on, the questions stop being about accuracy and start being about where the data goes.

  • Where the data may go

    For most organisations an approved model provider is acceptable, and that is the fastest route. For some records it isn’t, and then most vendor reference architectures stop being available on page one. We build either way.

  • The systems are inside

    The work worth automating lives behind a firewall, on hosts that accept no inbound connections and won’t be getting any.

  • Someone signs it off

    A reviewer will ask where data goes, who can decrypt it, and what gets recorded. Vague answers end the project.

What we do

  • Run the model where your policy allows

    Most engagements use an approved hosted model — most accurate, nothing to operate, and prompts reach that provider under your agreement with them. Where data genuinely cannot leave, local inference keeps everything on your hardware. Your decision per use case, not a constraint of ours.

  • Reach internal systems without opening ports

    Outbound-only tunnels and reverse SOCKS give the agent a route to private services. Nothing inbound needs to be exposed, so there is no new firewall rule to argue for.

  • Put it where the work happens

    Command line, desktop, mobile, and an end-to-end encrypted chat surface. Tool access to your own systems over the Model Context Protocol.

Built on our own stack

We don’t resell a hosted wrapper around someone else’s API. Delivery runs on kavilo — software we write and maintain ourselves, which is why we can change how it deploys to fit your constraints instead of asking you to fit a product.

Single binary
Statically compiled, with nothing else to install or run. Copies onto a host that has no internet access.
Providers
One interface over local inference (Ollama or vLLM) and hosted APIs (Anthropic, OpenAI, Gemini and DeepSeek), with retry and circuit breaking between them.
Tools
Filesystem, shell, web and retrieval tools, plus Model Context Protocol clients over stdio and streamable HTTP for reaching your services.
Messaging
A hub whose server never holds plaintext: X3DH and PQXDH key agreement, Double Ratchet, sender keys for groups, and sealed sender.
Networking
Outbound-only HTTP tunnelling, and bidirectional SOCKS5 over TLS or QUIC.
Source
Closed. Third-party inspection and source escrow can both be arranged on request, as priced options.

How engagements start

Three stages, and you can stop after any of them. Most clients begin with Assess, because it’s the cheapest way to find out whether we’re any good.

Assess

We map where your data can and cannot go, which systems hold the value, and what your reviewers will require. Fixed scope, fixed fee. It ends with a written architecture you own, whether or not you continue with us.

Deploy

We build it: topology of your choosing, connections to your internal systems, a surface your people actually use, and runbooks at handover.

Operate

Upgrades, model and provider changes, and acceptance tests re-run on every change so behaviour is measured rather than assumed.

And three ways to run it

Fully hosted by us when you want the outcome and no infrastructure — quickest to value, and your data either stays in your systems under scoped read-only access or is replicated to us, your choice. An approved hosted model in your environment when your policy already permits one, which is most organisations. Entirely on your hardware when data genuinely cannot leave. We will tell you which one you need, including when it is the cheap one.

What each stage includes →See it graded on seven industries →

Finish

Start with the boundary, not the model.

An Assess engagement is fixed scope and fixed fee, and ends with a written architecture you own — whether or not you continue with us.