Skip to content
TokenShunt
Platform & approach

The TokenShunt Method.

One job: stop your engineers' AI tools from paying frontier prices for routine work — without changing the tools, and without letting quality slip.

01 · Token map

Know where the tokens go.

Before anything changes, we classify your agents' real sessions by job and price each one. Most teams find the bill is dominated by reading, not reasoning.

  • Session and usage data from Claude Code, Cursor, and Copilot
  • Token use broken down by team, tool, and job type
  • Cost per job at current pricing
  • A baseline every later saving is measured against
Agent sessions → token mapIllustrative share of tokens by job · green = routable
  1. 01File readsWhole files pulled into context
  2. 02Repo scansSearch, grep, find-usages loops
  3. 03BoilerplateTests, types, config, scaffolding
  4. 04EditsChanges from a clear spec
  5. 05DecisionsArchitecture, debugging, novel problems

02 · Routing layer

Rules, not suggestions.

Telling an agent to use a cheaper model is a suggestion it can drop. The routing layer enforces the route outside the agent, where it can't be skipped.

  • Jobs classified before they reach a model
  • Hard limits, such as large files never reaching the frontier model
  • Per-repository policies on which models may see which code
  • Automatic fallback to the frontier model when a gate fails
  • Cache & repo index

    Repeated lookups, symbol and file search

    ≈ free
  • Cheap worker models

    Reading large files, repo scans, boilerplate, tests

    Lowest
  • Mid-size models

    Edits from a clear spec, small refactors

    Moderate
  • Frontier model

    Architecture, debugging, novel problems

    Highest

Spotify Engineering (September 2026) reported about 90% fewer Claude Code tokens in tests of a two-model routing setup on a Java monorepo. Spotify is not a TokenShunt customer. Your savings depend on how much of your token use is reading and boilerplate, which is why we measure your own sessions first.

03 · Worker models

Cheap models for routine work, wherever it has to run.

Reading files and writing predictable code don't need frontier intelligence. Worker models handle them at a fraction of the cost, via API or on your own hardware.

  • Chosen per job on measured quality and cost on your code
  • Commercial small models or open-weight models
  • Self-hosted when source code must stay on your network
  • Swapped as better or cheaper models ship
Engineer toolsClaude CodeCursorGitHub CopilotInternal agents
Routing layerJob classificationFile-size limitsPer-repo policiesFallback to frontier
Worker modelsSmall commercial modelsOpen-weight modelsSelf-hosted
ObservabilityCost per routeQuality gatesPer-team dashboardsRouting log
Your environmentCloud accountData centerIsolated network
Engineers keep their tools; everything below them runs in infrastructure you control and your platform team can own.

04 · Verification

Same output, proven on your code.

Routed and unrouted runs are compared on real tasks from your own repositories. A job stays on the cheap route only where it meets your bar.

  • Test set built from your own tickets and pull requests
  • Checks against your tests, builds, and linters
  • Blind spot-checks by your engineers
  • Per-job quality gates
  • Tests, builds & linters

    Broken or non-compiling changes

    Every routed change

  • Routed vs. unrouted runs

    Quality gaps per job type

    Before rollout · new models

  • Blind engineer spot-checks

    What automated checks miss

    Weekly sample

  • Cost tracking

    Savings drifting back up

    Continuous

Automated checks catch most problems immediately; engineer review confirms what automation can't. Any job that fails a gate goes back to the frontier model.

05 · Audit trail

Every routed change, traceable.

Each request records the job, the rule that routed it, and the model that handled it — so any change can be traced back when someone asks how it was produced.

  • Routing log exportable to your SIEM
  • Model recorded for every routed change
  • Documentation your risk and audit teams can review
  • Retention governed by your policy
Audit trail: every routed change, traced to its model
  1. Agent requestsession 41f2 · read src/billing/*
  2. Route decisionrule: file > 350 lines → worker
  3. Modelworker-small · v2026.09
  4. Changediff 7c1e · 3 files
  5. Pull requestPR #4812 · tests passing
  6. Reviewapproved · engineer on record

06 · Ongoing optimization

Savings that keep up with the models.

Model prices and capabilities change every few months. Routes are re-evaluated against your own tasks and moved only when the evidence supports it.

  • Cost and quality monitored per route
  • New models tested against your task set
  • Monthly savings reported against your baseline
  • Changes attributed to specific routing decisions
The optimization loop: measure, route, verify, monitor, rebalance — then repeat as models and prices change.MeasureRouteVerifyMonitorRebalanceEVERY MONTHCheaper.Still right.
New models and price changes are re-evaluated against your own tasks, and routes move only when the evidence supports it.

What makes us different

Not a model vendor. Not a generalist consultancy.

Model vendors and AI coding tools earn more when you use more tokens. Generalist firms sell frameworks. We have one job: make the routine work cheap and prove it stayed right.

Swipe to compare

Comparison of TokenShunt with typical model vendors and generalist consultancies
CapabilityTokenShuntTypical model vendorTypical generalist consultancy
Cuts AI coding spend without changing engineers' toolsIncludedNot their incentiveRarely
Vendor-neutral model choice per jobIncludedOwn modelsVaries
Hard routing rules, not prompt guidanceIncludedNot typicalNot typical
Self-hosted worker models where code can't leaveIncludedLimitedVaries
Quality verified on your own repositoriesIncludedPublic benchmarksLimited
Savings measured against your own baselineIncludedNot typicalVaries
Routing log for every changeIncludedVariesRarely
Runbook your platform team can ownIncludedDocs & supportLimited

Start the conversation

See the method on your own sessions.

Tell us which coding tools your engineers use and roughly what you spend. We'll show you where the tokens go and what routing would change.

We respond within 1 business day. Mutual NDA available before any data discussion.