Build a website
From a first professional site to Website 3.0 with CMS, multilingual workflows, automation and custom functionality.
View websites →westudiolab
Software Engineering & System Evolution
Understand problems. Build solutions. Evolve systems.
westudiolab develops digital systems, integrates AI and handles technical tasks for private individuals, self-employed professionals, companies and corporations. From a small intervention to a major transformation, what matters is that the solution fits the problem.
01 / Entry points[01] [02] [03] [04] [05] [06] [07] [08] [09]
How can we help?
Private individual, self-employed professional, small business or corporation: the right entry point depends on the task, not on the size of the client.
Google Ads
Set up, review or manage campaigns with a focus on real enquiries, calls and clean search terms.
View Google Ads →OpenAI Ads
Support for setting up audience-focused ads and campaigns in OpenAI Ads Manager, including marketing guidance, visuals and landing pages when needed.
View OpenAI Ads →AI Automation
Integrate AI into existing workflows and develop agents, assistants or automation for concrete tasks.
View AI →Apps & Vibe Coding
Develop web, smartphone and desktop apps, accelerate prototypes and move stalled projects forward.
View apps →Bookkeeping preparation
Prepare receipts, invoices and account movements in a structured way, with WESALDO in the background and ready for your accountant.
View bookkeeping →Find paper files digitally
WE-SORT links paper files with a database and QR codes so physical documents can be found quickly.
View WE-SORT →Audits
From a 30-minute website audit to reviews of complex systems and strategic initial consulting.
View audits →Robotics
Robotics consulting, software customization, a reception robot in planning and a showroom in Berlin in preparation.
View robotics →02 / Expertise
Decades of systems experience meet applied artificial intelligence
westudiolab is the technology and development division of WE Studio Media GmbH. It grew out of many years of work with software, digital production processes, and artificial intelligence. While WE Studio provides spaces and photo and video productions, the Lab develops digital systems, web apps, and AI-assisted tools, independently, but under one company roof.
Peter Gesierich
Founder & Lead Architect
40+ years of software development, from punch card systems and the Sinclair ZX Spectrum to today's web and AI infrastructure. Specialist in systems integration, pragmatic AI orchestration, and technical search optimisation. The focus is on understandable solutions that evolve established systems step by step while protecting business-critical connections. Authorised signatory of WE Studio Media GmbH and responsible for building its new technology division, westudiolab.
David Reger
DATA & AI PLATFORM ENGINEER
Specialist for data platforms, AI integrations, and scalable system architectures. Focus on building the data foundation for productive AI applications, from data integration and RAG systems to agentic workflows. Expertise in Python (FastAPI, SQLAlchemy), SQL/database engines, Kubernetes, and modern data stacks. Experience with web scraping, PDF parsing, ERP integrations, semantic search, and event-driven architectures. Technical focus: Lakehouse platforms, streaming systems, data processing, LLM orchestration, and production-ready AI deployments. From data sourcing to the robust deployment of AI-enabled services.
03 / Product & System Strategy
Evolve what exists. Build new products on solid foundations.
Strategy does not begin with a programming language. We consider goals, users, workflows, and existing value, then shape a solution that works today and can continue to grow tomorrow.
[01]
Purpose and value first
We clarify which problem the application should solve, who will use it, and which workflows genuinely need improvement. Only then do we decide on features and technology.
[02]
Preserve what matters
In established systems, we protect user accounts, data, interfaces, and working processes. For us, modernisation means thoughtful evolution rather than a premature complete rebuild.
[03]
Build new products lean
New websites, web apps, and AI applications are designed to be modular and extensible. The architecture follows the project rather than a preferred programming language.
[04]
Grow step by step
We deliver verifiable stages, test against real requirements, and expand only where measurable value emerges. This keeps the product understandable, manageable, and sustainable.
Technical process for legacy modernisation
For technical decision-makers, this is our proven five-phase model. It keeps change controlled, reversible, and as unobtrusive as possible for existing users.
[01]
Phase 1: Visibility & Security
Mapping the system landscape, processes, and dependencies. Establishing monitoring and blue-green deployment strategies to ensure every change remains reversible.
[02]
Phase 2: Modularise, Don't Destroy
Structuring existing code into a "Modular Monolith". Introducing clean APIs between modules instead of "Wild West" data access.
[03]
Phase 3: Strangler Approach & Facade Architecture
Implementing the Strangler Fig Pattern: New features are built externally, legacy endpoints are replaced incrementally without users noticing the transition.
[04]
Phase 4: Data Migration Without Shock
Utilising techniques like Change Data Capture and read replicas. Parallel operation of old and new data paths with constant validation.
[05]
Phase 5: Decommission & Cleanup
Targeted shutdown of legacy components and removal of dead code. Handover of a lean, modern core system to your team.
04 / Technology & Data Sovereignty
Technology follows the project, not the other way around.
We combine proven software engineering with the purposeful use of artificial intelligence. The goal is not the longest technology list, but a solution that remains understandable, maintainable, and under your control.
[01]
Websites 3.0
A digital presence becomes an intelligent application: it connects content, data, and processes, assists users with AI, and can automate useful workflows without surrendering control or data sovereignty.
[02]
Architecture that fits
We selectively evolve existing technologies or choose a stack for new products that fits usage, team, and growth. What matters is not the programming language, but a durable and maintainable solution.
[03]
Controlled infrastructure
Our own AI infrastructure on NVIDIA DGX Spark and a Hetzner server location in Nuremberg enable controlled, privacy-conscious operating models. Depending on the project, sensitive data can be processed locally or in Germany.
Technology in detail
For technical teams: the concrete selection always follows the task, the existing environment, and the preferred operating model.
[01]
Backends & APIs: Python (FastAPI, SQLAlchemy, Pydantic), Node.js (Next.js, Express), REST APIs, GraphQL, and WebSockets for real-time integration.
[02]
AI & LLMs: RAG pipelines, agentic workflows, multi-model orchestration (OpenAI, Anthropic, and local models via Ollama), LiteLLM Gateway, semantic search, and embeddings.
[03]
Data & integration: relational and document databases, lakehouse architectures, web data extraction with Playwright and BeautifulSoup, PDF parsing, ERP/CRM integration, and event streaming with Kafka.
[04]
Frontend & SEO: Next.js with App Router and Server Components, React, Tailwind CSS, technical SEO, structured data, accessibility, and Core Web Vitals.
[05]
DevOps & infrastructure: Docker, Kubernetes, CI/CD with GitHub Actions, Ubuntu Linux, Caddy reverse proxy, systemd, monitoring, NVIDIA DGX Spark with Ubuntu-based DGX OS for local AI workloads, and Hetzner hosting in Nuremberg.
05 / Lab & Collaboration
Technology isn't created only in the browser.
Alongside software and AI, we test prototypes, physical systems and new connections between digital control and the real environment. Robotics is being given its own section on westudiolab for this purpose.
Prototyping & physical systems
Sensors, control, interfaces and AI-supported interaction for concrete experiments and new applications.
Local AI & Infrastructure
Our own computing infrastructure allows us to test models, agents and data paths in a controlled way and combine them for specific applications.
Collaborations
For suitable projects, we work with companies, developers and technical teams on a project basis with a clear division of tasks.
Robotics in Berlin
Robotics consulting, reception robots in planning, and a showroom in Berlin in preparation.
To the Robotics section →Insights from the Lab
Using local AI where data needs to remain protected
Not every AI application belongs in a public cloud without careful consideration. At westudiolab, we evaluate local models and clearly defined interfaces for tasks where control over data, cost, and operations matters.
The aim is not to use the largest possible technology, but to build a solution that fits the real process and whose limitations remain understandable.
Modernising legacy systems without losing existing users
A clean-sheet rebuild may sound simple, but it often puts accounts, permissions, data relationships, and familiar workflows at risk. We therefore begin by understanding real dependencies and replacing parts step by step.
This allows a new, lean system to emerge while important connections are transferred and verified in a controlled way.
Central content instead of separately maintained language versions
Multilingual websites should not consist of separate text copies that quickly drift apart. Our content models begin with one centrally maintained German version; the English version is generated automatically and can be reviewed editorially.
We apply the same principle to new web apps: one reliable source, multiple suitable outputs.
When something needs to be solved.
Describe the problem, the existing system or the idea. It may be a clearly scoped technical task, a new product, a complex modernisation or a possible collaboration. We will assess which next step makes sense.