What INWT Statistics does

INWT Statistics GmbH is a Berlin-based data science consulting firm founded in 2011 that delivers end-to-end "Full Stack Data Science" services to enterprise clients, organized around three service packages: Proof of Concept (PoC) for feasibility assessment (~20% of work), MVPs and Data Products for production deployment and ongoing SLA support (~40%), and Legacy zu Goldstandard for modernizing aging code and infrastructure (~40%). The firm serves medium-to-large enterprises across Germany and selectively internationally, with a 22-person team led by Managing Directors Dr. Amit Ghosh and Mira Céline Klein alongside co-founder Prof. Dr. Steffen Wagner.

Technically, INWT combines predictive analytics, software engineering, and MLOps into a single delivery model. The predictive analytics stack spans Bayesian statistics, gradient boosting (XGBoost), foundation models (TabPFN), mixed-effects and survival models, RAG systems (Langchain, Qdrant), and LLM fine-tuning/serving using Torchtune, Transformers, Ollama, and vLLM across OpenAI, Llama, Mistral, and DeepSeek models. Multimodal capabilities include combining tabular features, free text, and images via fine-tuned GPT-4o. The software layer delivers APIs, dashboards (Shiny, Redash, Vue.js), data warehouses, and ETL pipelines, while the operations layer standardizes on AWS/Azure/Hetzner, Docker, Kubernetes, and CI/CD. INWT also maintains an open-source R package for Kafka integration and an internal Predictive Analytics Cube (PAC) methodology for standardized model benchmarking.

The business model is purely professional services: project-based consulting engagements sold directly to enterprise buyers, with no public pricing, no self-serve product, no marketplace, and no disclosed outside funding. Customer acquisition relies on thought leadership through a technical blog (active since 2012, 80+ posts), white papers, podcasts, LinkedIn, and an English-language mirror site supporting international outreach. The named client base spans 13+ enterprise logos across e-commerce (Zalando, Otto), automotive (Sixt), media (Spiegel, Zeit Online, dpa, Ströer), healthcare and research (Charité, MPI), fintech (Smava), gaming (Tipico), energy (BWE), and environmental monitoring (AQI), with additional engagements in IoT (Wattro), ESG reporting, and election forecasting.

Name

INWT Statistics

Legal name

INWT Statistics GmbH

Website

https://inwt-statistics.de

Company type

Private

Founded year

2011

Operating status

Operating

Headcount range

11–50 employees

Short description

INWT Statistics GmbH is a Berlin-based data science consulting firm founded in 2011, delivering end-to-end predictive analytics, software engineering, and MLOps services to enterprise clients across the DACH region through PoC, MVP, and legacy-modernization engagements.

Ownership category

akta.pro rank

Product category

Data Science Consulting

NAICS

Custom Computer Programming Services (541511), Computer Systems Design and Related Services (5415)

SIC

Services-Computer Programming, Data Processing, Etc. (7370), Services-Computer Programming Services (7371), Services-Engineering, Accounting, Research, Management (8700)

akta.pro primary industry

End-to-End MLOps & ML Platform Suites (HDAAABAA)

akta.pro secondary industries

Marketing Analytics & Measurement Strategy (Attribution, MMM, KPIs) (MPAAAFAH), Data Quality, Profiling & Validation (within Integration) (HDAEACAI)

Keywords

• Data science consulting
• Predictive analytics services
• Machine learning consulting
• MLOps implementation
• Data engineering services

Headquarters

HQ city

Berlin

HQ country

Germany

HQ region

Europe

Offices1 record

Markets served

GTM type

B2B

Offering type

Services

Cost components

Personnel, Technology or R&D, Infrastructure, Marketing or Sales, Operations

Revenue model

1. Professional Services / Consulting: INWT operates as a data science consulting company offering three service packages: (1) Proof of Concept (PoC) for feasibility assessment and prototype development, (2) MVPs and Data Products for production-ready ML solutions with ongoing support and SLA, and (3) Legacy to Goldstandard for code modernization, refactoring, cloud migration, and feature extension. Revenue is generated through consulting engagements and project-based work for enterprise clients.

Go-to-market motion1 record

Distribution channels2 records

Marketing channels5 records

Core offering

INWT Statistics is a data science consulting firm that delivers Full Stack Data Science across the lifecycle from prototype to production. The company sells three consulting service packages: Proof of Concept (PoC) for feasibility assessment and prototyping, MVPs and Data Products for production deployment with MLOps and SLA support, and Legacy to Goldstandard for code refactoring, cloud migration, and feature expansion. Underpinning these packages are three capability areas: Predictive Analytics (Bayesian statistics, gradient boosting, TabPFN, LLM fine-tuning), Software Development (APIs, dashboards, ETL pipelines), and Operations/MLOps (Kubernetes, Docker, CI/CD on AWS/Azure/Hetzner).

Product overview

INWT Statistics is a data science consulting firm (founded 2011) offering three core consulting service packages: Proof of Concept (~20% of work) for feasibility assessment and prototyping; MVPs und Datenprodukte (~40%) for production deployment and ongoing support; and Legacy zu Goldstandard (~40%) for code modernization. These packages are supported by three technical capability areas: Predictive Analytics (custom forecasting with Bayesian methods, foundation models, gradient boosting, LLM fine-tuning); Software Development (APIs, dashboards, data warehouses, ETL pipelines); and Operations/MLOps (CI/CD, Docker, Kubernetes, cloud infrastructure). The company also develops open-source tools including the R-Paket kafka for Kafka integration and maintains proprietary methodologies like the Predictive Analytics Cube for model comparison.

Differentiator

Problem solved

Functional benefit

Products and services

• Proof of Concept (PoC) A consulting service package that builds a prototype of a forecast model to determine feasibility and value. Includes data validation and analysis, reusable code, leverage assessment, technology recommendations, and a production roadmap for medium and large enterprises considering ML or predictive analytics investments.
• MVPs and Data Products A consulting service package that brings data products and ML prototypes into production and provides ongoing support. Covers development, quality assurance, MLOps, operations, and Support & SLA for enterprise clients running production ML systems.
• Legacy to Goldstandard A consulting service package that makes aging code maintainable and adapts it to current standards. Includes refactoring, technical debt reduction, scaling, modularization, cloud migration, and feature expansion for organizations with legacy ML or data science codebases.
• R-Paket kafka An open-source R package enabling Apache Kafka integration for real-time data streaming. Allows R developers to produce and consume Kafka messages via producer/consumer classes, demonstrated with a real-time ISS location tracking use case.

Quantifiable outcome

• Improved RFID device recognition accuracy from 90% to 99% for Wattro GmbH
• +4 more outcomes

Named customers16 records

Segments1 record

Ideal customer profiles1 record

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

AI capability11 records

Feature8 records

Partnerships

Two partnerships are on record, tiered core and minor.

• LOT InternetcoreStrategic or Co-development PartnerLOT Internet provided access to a dataset of 8000+ vehicles listed on mobile.de for predictive analytics benchmarking. This partnership enabled INWT's research comparing LLM-based forecasting (GPT-3.5, GPT-4o, Llama, Mistral) with XGBoost on real-world tabular and text data for used car price prediction.
• Nature JournalminorOthersTabPFN, a foundation model for tabular data that INWT implements and benchmarks, was published in Nature (January 2025). This is a reference to an academic publication, not a formal partnership.

Scale indicators5 records

Recent moves6 records

Expansion highlights5 records

Direct peers

• eoda GmbH: Kassel-based data science consultancy established in 2010, with comparable offerings in predictive analytics, R/Python development, and BI/dashboarding for German mid-market and enterprise customers.
• Alexander Thamm GmbH: Founded 2012 in Munich, one of the largest German data science and AI consultancies serving DACH enterprise clients across similar verticals. Closest comparable on business model, size trajectory, and 'Full Stack Data Science' positioning.
• statworx: Frankfurt-based data science, ML, and AI consultancy founded in 2013 with a similar boutique-academic profile, offering consulting, academy training, and predictive analytics services to enterprise clients.

Emerging players

• H2O.ai: AI cloud platform with autoML and LLM capabilities plus a services practice; targets similar predictive-analytics use cases and could either compete with or complement INWT's consulting engagements.

Others

• Databricks: Data and AI platform with a large professional-services arm; partially competes where INWT implements data lakes, ETL, and ML platforms, and is also a potential tooling partner for INWT's MLOps engagements.

Broad incumbents

• Capgemini Engineering: Global engineering and digital services provider with strong DACH presence and AI/data engineering offerings; competes on large enterprise modernization and MLOps implementation work.
• McKinsey QuantumBlack: McKinsey's advanced analytics arm delivering ML and AI transformation engagements; overlaps on enterprise predictive-analytics work and targets the same C-level buyers INWT reaches via its marquee clients.
• Deloitte AI & Data: Global consulting practice delivering data, analytics, and AI services across all major industries; competes with INWT for large enterprise transformation mandates that bundle predictive analytics with broader advisory work.
• BCG GAMMA: BCG's dedicated data science and AI unit providing enterprise analytics, ML, and MLOps consulting globally; competes for the same German enterprise predictive-analytics mandates from a much larger scale.
• IBM Consulting (Data & AI): Global systems integrator offering enterprise data science, watsonx, and MLOps services; competes especially in regulated verticals (healthcare, public sector) where Charité- and MPI-style clients operate.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat4 records

Key risks6 records

Key highlights7 records

Customer concentration

• GitHubgithub.com/INWTlab
• Xx.com/inwt_statistics

Revenue estimate

Valuation estimate

Number of profiles

Profiles3 records

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about INWT Statistics