Data & AI Engineering · Made in Germany
Production-grade data
and AI platforms.
We design, build, and run the data and AI infrastructure that German Mittelstand and enterprise companies depend on. A hands-on team of seven, GDPR-native by default, based in Ingolstadt.
Built on the tools your team already runs
What we build
Our technical surface area.
Production data and AI services, from robust pipelines to live ML systems.
Data engineering & ETL
dbt, Airflow/Dagster, Kafka. Pipelines that survive real workloads.
MLOps & model serving
CI/CD for ML, drift monitoring, scalable serving on Triton, SageMaker, Vertex AI.
Analytics & BI
Snowflake, BigQuery, Redshift, Databricks. Modelled, governed, fast.
Cloud & platform
Terraform IaC, security hardening, FinOps across AWS, Azure, GCP.
LLM & RAG integrations
Production RAG, vector databases, retrieval evaluation, agentic workflows.
Custom ML models
Deep learning, time-series, NLP when off-the-shelf isn't enough.
Healthcare AI
EEG analysis, imaging pipelines, GDPR/MDR-compliant clinical workflows.
Architecture & advisory
Architecture reviews, roadmaps, tooling selection.
Engage
Pick the engagement that fits.
Productised entry points with transparent pricing. From a single day of engineering time through to a 12-month embedded team.
1 day · single session
Architecture Day
from €2,500
A focused day with an engineer from our team on one specific question: architecture review, vendor selection, scaling problem, compliance gap. Written follow-up included.
- One specific question, deeply answered
- Same person scopes and delivers
- Written summary within 48h
3 weeks · fixed price
Data & AI Readiness Audit
from €15,000
We audit your architecture, MLOps maturity, cloud cost, and compliance posture. You get a written report, a 90-minute executive readout, and a 12-month roadmap your board can sign off on.
- Architecture and cost review
- Compliance gap analysis
- 12-month roadmap
8–16 weeks · milestone-billed
Production Platform Build
from €60,000
We embed 2 to 4 engineers from our team with yours and ship a production system: platform modernisation, MLOps, RAG/LLM, compliance hardening. Fixed scope, fixed price.
- Fixed scope, fixed price
- 2 to 4 engineers embedded
- Production system plus handover
6–12 month retainer
Embedded Team
on request
A fractional data and AI platform team. Ongoing reliability, new feature delivery, on-call MLOps, without the long hiring cycle for platform engineers.
- 1 to 3 engineers from our team
- Ongoing reliability plus features
- On-call MLOps coverage
Something else in mind? Tell us about your project, we scope custom engagements where it makes sense.
Beyond consulting
We also build and host products.
Two pieces of the stack we run for ourselves and for clients.
How we work
Your partner in data and AI transformation.
- Production first.
- Pipelines, model serving, SLAs that hold up in real use.
- Small team, direct delivery.
- The person who scopes the work is on the delivery. No offshore handoffs.
- Fixed scope by default.
- Discovery, pilot, production, with documentation and handover.
- GDPR-native architecture.
- Least-privilege IAM, private networking, auditability by design.
- Healthcare-grade trust.
- EEG, imaging, clinical workflows under GDPR and MDR. The same bar applies to every client.
Whether you're standardising ELT with dbt, rolling out model serving on Triton or Vertex, or integrating LLMs with retrieval, we align architecture and delivery to your stack and KPIs.
Leadership
Engineering-led.
NeuralMedic is led by Jonas Heinzmann, computer scientist and founder. The company is built on a single principle: data and AI systems should run in production, not sit in slides.
Based in Ingolstadt, Bavaria. The team includes full-time engineers and working students in computer science and related fields. We say no to work we can't deliver.
FAQ
Questions, answered.
We build reliable data pipelines (ETL/ELT), production-grade MLOps and model serving, analytics/BI platforms, secure cloud infrastructure, and LLM integrations (RAG and vector databases) for enterprises.
dbt, Airflow/Dagster, Kafka; Snowflake, BigQuery, Redshift, Databricks; AWS, Azure, GCP; model serving on Triton, SageMaker, and Vertex AI; vector databases and retrieval evaluation for LLMs.
We start with discovery and scoping, then propose a plan with milestones. Delivery is iterative with regular demos, followed by documentation, handover, and optional ongoing support.
POCs: 2 to 4 weeks. Production implementations: 4 to 12 weeks depending on scope, integrations, and compliance requirements.
GDPR-native data handling, least-privilege IAM, encryption in transit and at rest, private networking, and auditable pipelines. We follow SOC 2-friendly practices and sign DPAs when required.
Fixed-scope (SOW) or time-and-materials depending on uncertainty. We run a short discovery, provide transparent estimates, and optimise for ROI and cost control.
Yes. We sign DPAs, deploy within your VPC or on-prem, enforce least-privilege access, and ensure no data leaves your environment without approval.
Ready to ship?
Tell us about your project.
The fastest path is a 30-min architecture review on the calendar. Or send us a few lines about what you're building; we respond within one business day.
Book a 30-min architecture review- info@neuralmedic.de
- Phone
- +49 (0) 15203570876
- Office
- Haslangstraße 49
85049 Ingolstadt, Germany