Extensions

Services to evaluate, optimize, and maintain your AI setup.

Enhance your AI ticket system with specialized extensions that add new capabilities, ensure ongoing performance, and maintain system health. Our extension services include custom modules, performance monitoring tools, quality assurance frameworks, and maintenance packages. We provide regular updates, bug fixes, security patches, and feature enhancements to keep your AI system running at peak efficiency.

Fine-tuned Model

Custom-trained model artifact

from 4 000 €1-2 weeks
generic
fine-tuningmodeldeployment

Who it’s for

  • You already have a dataset and want a deployable model artifact

Deliverables

  • Fine-tuned model artifact
  • Benchmark summary

Prerequisites

  • Dataset available in agreed format
  • Target runtime constraints (CPU/GPU, latency)

Included

  • One training run + benchmark

Excluded

  • Dataset creation/labeling

Editions

Fine-tuning · 4 000 €

Process

1Input Review
  • •Verify dataset format
  • •Confirm constraints
2Training
  • •Fine-tune model
  • •Run benchmarks
3Delivery
  • •Deliver model artifact
  • •Provide benchmark summary
4Optional Integration
  • •Integration is separate service

FAQ

Request fine-tuned modelBook 15-min call

Evaluation Report

Model performance insights, statistics, and recommendations

from 4 000 €3-7 days
generic
evaluationmetricsinsights

Who it’s for

  • You need clear statistics on model accuracy, weak spots, and failure modes
  • You want actionable insights before investing in tuning, fine-tuning, or production rollout

Deliverables

  • Performance report with accuracy statistics, precision/recall/F1 where relevant, and confidence insights
  • Error analysis with examples, patterns, and prioritized recommendations
  • Management-ready summary of what to improve next

Prerequisites

  • Model outputs on a labeled set or evaluation dataset
  • Business priorities for accuracy, latency, and risk tolerance

Included

  • One review call for the simple case

Excluded

  • Model retraining (separate service)

Editions

Simple Case · 4 000 €Complex Case · 10 000 €

Process

1Inputs
  • •Collect predictions + labels
  • •Confirm target metrics and decision thresholds
2Analysis
  • •Compute metrics
  • •Identify failure modes
3Report
  • •Deliver statistics, insights, and recommendations
  • •Highlight accuracy and performance improvement opportunities
4Review
  • •Short walkthrough

FAQ

Request evaluationBook 15-min call

System Tuning

Optimize an existing model for best accuracy and runtime performance

from 4 000 €3-7 days
generic
optimizationtuningaccuracy

Who it’s for

  • You already have a running model and want the best possible accuracy, confidence behavior, and runtime performance
  • You want targeted optimization without starting a full new model-development project

Deliverables

  • Accuracy and performance tuning plan for the existing model
  • Optimized thresholds, prompts/configuration, runtime settings, or model parameters within scope
  • Before/after tuning report with accuracy, latency, and reliability observations

Prerequisites

  • Access to the current model, evaluation results, representative examples, and runtime constraints

Included

  • One tuning iteration within scope

Excluded

  • Full model retraining or a new model architecture unless scoped separately

Editions

Model Optimization · 4 000 €

Process

1Review
  • •Assess current accuracy, latency, confidence behavior, and failure cases
  • •Confirm target metrics and runtime constraints
2Tuning
  • •Optimize model configuration, thresholds, prompts, or runtime settings
  • •Improve robustness on known edge cases
3Validation
  • •Re-run evaluation and performance checks
  • •Document before/after results
4Handover
  • •Provide tuning report
  • •Next steps

FAQ

Request system tuningBook 15-min call

Ongoing Updates

Monthly model maintenance and improvement

from 2 500 €/monthmonthly
generic
subscriptionupdatesmaintenance

Who it’s for

  • You want a running model to stay accurate as tickets, processes, and business priorities change
  • You need a predictable monthly cycle for small improvements, quality checks, and release notes

Deliverables

  • Monthly model, prompt, dataset, or configuration updates within scope
  • Release notes and change summary
  • Lightweight evaluation after each update

Prerequisites

  • Change log of observed issues, new examples, and workflow changes
  • Access to recent samples or evaluation set

Included

  • 1 planned update cycle per month

Excluded

  • Large new feature development
  • Full model retraining unless scoped separately

Editions

Monthly Maintenance · from 2 500 €/monthManaged Improvement · from 5 000 €/month

Process

1Intake
  • •Collect changes
  • •Prioritize
2Update
  • •Apply changes
  • •Run checks
3Evaluate
  • •Quick evaluation
  • •Document impact
4Release
  • •Deliver update
  • •Release notes

FAQ

Request ongoing updatesBook 15-min call