TokenSchmiede

How I can help

A builder when you need one. A partner for your team.

Four ways to work together, shaped around your product, people and constraints. We agree on scope and a useful first milestone before implementation.

AI product engineering

Software kitchen

A good fit when: You have a workflow to improve or a product to build, and need someone who can own the technical delivery.

We define the smallest useful version, build it around the actual workflow and work through the operational details. Dictation, coaching applications and agent tooling are part of my own product work.

What we can work toward

  • A scoped prototype tested against your use case
  • Application and integrations, with tests where they matter
  • Deployment, documentation and a clear handover
Meet fape

Embedded AI engineering

A player-coach in your team

A good fit when: Your team wants to use AI coding agents more effectively while continuing to deliver reliable software.

I join the work as a senior or lead engineer. We improve context, task breakdown, review and verification while shipping features together. The learning happens in your codebase.

What we can work toward

  • Hands-on engineering in your existing stack
  • Agent workflows with review and verification practices
  • Pairing, documentation and knowledge transfer

AI security & hardening

A stronger existing stack

A good fit when: You operate a web application and need a clear view of its risks and a path to remediation.

We agree on the systems and scope, review architecture, access controls, dependencies and deployment, then prioritize fixes. This is defensive engineering for systems you operate.

What we can work toward

  • A prioritized review with actionable findings
  • Agreed fixes in application and configuration
  • Operational guidance for updates and access control

Regulated & on-prem applications

AI where your data belongs

A good fit when: Your use case involves sensitive documents, internal systems or restrictions on where data may be processed.

We start with data flows and hosting requirements, then choose models and integrations to fit. My background includes private equity and invoice automation at HiQ. Deployment choices follow your requirements.

What we can work toward

  • An architecture with explicit data and access boundaries
  • A prototype evaluated on representative tasks
  • A deployment plan for your infrastructure and review process

Working together

Clear scope. Tangible progress.

  1. 01

    Talk through the problem

    We discuss the workflow, existing stack, constraints and what success should mean.

  2. 02

    Agree on a first step

    A focused review, a prototype or an embedded engagement. Scope, responsibilities and deliverables are explicit.

  3. 03

    Build, review, hand over

    You see the work as it develops. We verify the result and leave your team with usable documentation.

Start a conversation

What would you like to make work better?

Tell me what you’re building, where you’re stuck and what a useful outcome would look like. We can work out the next step together.

Let’s talk about your project

Direct contact with Sebastian. English or German.