01
Find the bottleneck
We identify tasks where manual work, search effort, data quality, or response time creates real cost.
Software-first AI agency for production AI systems
AI consulting & software implementation for production AI systems.
1tm builds AI systems that work inside real business processes: RAG systems for company knowledge, AI chatbots, local LLMs, AI agents, automation, and custom software.
- We find workflows where AI can remove real friction
- We check data, risk, and integration effort before implementation
- We build software people can actually use, not isolated AI demos
Trusted context
20+ years in software engineering
German GmbH
SME and B2B focus
Privacy-aware implementation
01
Services for companies that need more than an AI demo
We combine strategic AI consulting, software engineering, data integration, and operation into systems that fit existing workflows.
AI consulting & strategy
Clarify use cases, data readiness, risk, and integration effort before budget flows into the wrong pilot.
View serviceRAG systems
Build search and answer systems that make company knowledge usable from approved sources.
View serviceAI chatbots
Build chatbots for websites, support, sales, and internal knowledge with clean process handoffs.
View serviceLocal LLMs
Evaluate when local, cloud, or hybrid model architectures make sense for privacy, quality, and cost.
View serviceAI agents & automation
Automate recurring research, document, and analysis workflows with controlled agentic systems.
View serviceCustom AI software
Build web apps, portals, data integrations, and AI features that fit real business software.
View service02
From bottleneck to production AI system
We do not sell abstract AI strategy. We work through the practical sequence: isolate the problem, test whether it is worth solving, reduce the risk, build the solution, improve it in operation.
02
Check feasibility and value
We assess the business case, data readiness, technical dependencies, privacy, and integration effort.
03
Build the pilot
We validate the business case with a focused prototype or MVP before the idea becomes a larger project.
04
Integrate into systems
We bring the solution into existing workflows, data sources, web applications, and operational processes.
05
Improve impact
After launch, we improve quality, adoption, and economics based on real usage.
03
Where this way of working already helps
These examples are not technology showcases. They show recurring patterns: structure work, make knowledge available, and accelerate decisions.
Product descriptions for online shops
Context
Many items need consistent copy; manual creation slows assortment and content operations.
Solution
We built an AI-enabled workflow for automated, search-friendly product description creation.
Result
Recurring copy work became a faster, more consistent, and easier-to-integrate process.
Chatbots for support and websites
Context
Customers and prospects expect fast orientation while knowledge is spread across websites and internal sources.
Solution
We designed chatbot solutions that can use verified external and internal information in a controlled way.
Result
Answers become more consistent, easier to find, and better embedded in the customer journey.
News analysis for sales signals
Context
Relevant market and news signals appear continuously, but manual monitoring is hard to run reliably.
Solution
We implemented an AI-supported web application for topic monitoring and automated reports.
Result
Manual research became a structured, repeatable analysis workflow for sales and market observation.
04
Products, platforms, and initiatives from 1tm
We don't just advise — we build. These are the products, client projects, and initiatives 1tm works on.
Own product
KI-ProduktInfo
From 5 hours to 15 minutes of product-data work a day.
AI-powered product-data automation for e-commerce: extract attributes, categories, and descriptions from images and sources, ready to export to Navision/Dynamics NAV.
ki-produktinfo.deOpen website
Client project
Datenmassiv
AI enablement for the midmarket.
As implementation partner, 1tm supports Datenmassiv with strategy and engineering — from audits and pilots to durable integration.
datenmassiv.comOpen website
Platform
Cosmic Council
One question, many minds, one wiser answer.
A platform for shareable teams of AI agents: specialized agents deliberate together and a chair synthesizes their perspectives into one answer.
cosmiccouncil.spaceOpen website
In development
HeartbeatGuardian
One tap a day to say 'I'm okay.'
In development: daily well-being check-ins for seniors — one tap a day gives family peace of mind without constant calls.
heartbeatguardian.comOpen website
05
Why AI projects depend on implementation, not just models
The model is rarely the whole problem. The real question is whether data, workflows, privacy, adoption, and operation fit together.
Privacy and sovereignty
Before sensitive information is processed, we clarify data flows, model access, and integration points.
Respect existing IT
We build solutions so they fit current systems, teams, and workflows.
Commercial function
Every implementation needs a clear business function, not just an impressive demo.
Work with domain teams
The best technical approach combines software expertise with the process knowledge of your domain teams.
06
For companies that need more than advice
1tm combines AI consulting, software architecture, web development, and B2B process understanding into a team that can deliver.
Martin Schubert
CEO | AI consulting, development, and business strategy
Develops AI strategies and technical solutions that fit real business workflows.
LinkedIn
Toni Thomä
CTO | AI engineering and software architecture
Turns complex AI technology into applications people can actually use.
LinkedIn
Markus Wahl
Software development, marketing, and B2B/e-commerce
Connects implementation work with product, market, and communication needs.
LinkedIn
Thomas Wild
E-procurement, SAP, and B2B solutions
Contributes experience in procurement workflows, SAP, and e-commerce integration.
07
Questions to settle before an AI project starts
Good AI projects do not start with a tool. They start with clear decisions about purpose, data, integration, and operation.
How do we identify an AI project worth building?
A useful AI use case reduces specific process costs, accelerates recurring decisions, or makes existing knowledge more reliable. We assess the bottleneck, data readiness, risk, and integration effort first.
Do our data need to be perfectly prepared?
No. Many projects start with a realistic data assessment. The important question is which data are needed for the first valuable use case and how quality, access, and privacy can be controlled.
Does 1tm only build prototypes, or production systems too?
Our focus is implementation. A pilot validates value and feasibility. After that, we integrate the solution into existing web applications, data sources, portals, or operational workflows.
Which companies are the best fit?
We mainly work with mid-sized B2B companies that have a clear process problem, existing systems, and the goal of bringing AI into daily operations in a controlled way.
For internal orientation
These materials help teams talk about AI potential internally. The real assessment starts with your workflows, data, and systems.
08
Find out where AI makes commercial sense in your business
Briefly describe your company, one workflow, or a concrete challenge. We will assess whether it should become an AI use case, a pilot, or first a better foundation.