# Tensorhead AB — Full Profile

> AI/ML engineering and consulting for startups and scaleups. Based in Gothenburg, Sweden. Available worldwide.

## About Tensorhead

Tensorhead AB is a freelance AI/ML consulting firm founded by Patrik Larsson. With 10+ years of experience across the full machine learning lifecycle — from data engineering and model development to MLOps and deployment — Patrik brings a practical, results-oriented approach to every project.

Whether you're building your first ML pipeline or scaling an existing system, Tensorhead can help you move faster and make better technical decisions.

**Company:** Tensorhead AB
**Founder:** Patrik Larsson, AI/ML Engineer
**Location:** Gothenburg (Göteborg), Sweden
**Availability:** Worldwide, remote-first
**Company registration:** Org.nr 559438-7853, VAT SE559438785301

## Services in Detail

### ML Model Development
Custom machine learning models designed for your specific business problems — from prototyping to production-ready solutions. From demand forecasting to computer vision, models are tailored to your data and constraints.
- **Typical stack:** PyTorch, Hugging Face, cloud-native training pipelines on AWS, GCP, or Azure
- **Use cases:** Demand forecasting, computer vision, NLP, recommendation systems, anomaly detection, classification

### MLOps & Infrastructure
Scalable ML pipelines, model serving, monitoring, and CI/CD — so your models run reliably in production. Automated training, model registries, and monitoring dashboards on AWS or Kubernetes — so your team can ship models without firefighting.
- **Typical stack:** MLflow, Kubernetes, Docker, Terraform, GitHub Actions, AWS SageMaker
- **Use cases:** Model deployment automation, A/B testing infrastructure, model monitoring, retraining pipelines

### LLM & GenAI Solutions
RAG systems, fine-tuning, prompt engineering, and agent workflows — leveraging the latest in generative AI. Whether it's a customer-facing chatbot, an internal knowledge assistant, or an autonomous agent pipeline — GenAI systems that work reliably at scale.
- **Typical stack:** LLM APIs (OpenAI, Anthropic, open-source), LangChain/LangGraph, vector databases, RAG frameworks
- **Use cases:** Customer support chatbots, knowledge assistants, document processing, code generation tools, autonomous agents

### Data Engineering
Data pipelines, feature stores, and data quality frameworks that form the foundation for successful ML projects. ETL pipelines, data validation, and feature engineering.
- **Typical stack:** Apache Spark, dbt, Airflow, BigQuery, Snowflake, Delta Lake
- **Use cases:** ETL/ELT pipelines, feature stores, data quality monitoring, data lake architecture

### AI Strategy & Advisory
Technical due diligence, roadmap planning, and hands-on guidance to help you make the right AI investments. Evaluate build-vs-buy decisions, assess vendor tools, and create realistic AI roadmaps grounded in technical feasibility.
- **Use cases:** Technical due diligence for investors, AI roadmap creation, vendor evaluation, team capability assessment

## Full Technology Stack

**ML & AI:** Python, PyTorch, Hugging Face, scikit-learn, XGBoost, TensorFlow
**GenAI:** LLM APIs, RAG, AI Agents, LangChain/LangGraph, Vector DBs (Pinecone, Weaviate, pgvector)
**Cloud & Infra:** AWS, GCP, Azure, Terraform, Docker, Kubernetes
**Data:** Apache Spark, dbt, Airflow, BigQuery, Snowflake
**APIs & Serving:** FastAPI, MLflow, BentoML
**Monitoring:** MLflow, Weights & Biases, Grafana

## Engagement Process

1. **Discovery** — Learn about your business, data, and goals. Identify where AI can make the biggest impact.
2. **Proposal** — A clear scope, timeline, and approach — so you know exactly what you're getting before we start.
3. **Build** — Iterative development with regular check-ins. You see progress early and often.
4. **Deliver & Support** — Production deployment, documentation, and knowledge transfer. Support continues until everything runs smoothly.

## Frequently Asked Questions

**Q: What types of AI projects does Tensorhead handle?**
A: Everything from custom ML model development and MLOps infrastructure to GenAI/LLM solutions and data engineering. Projects range from building a first ML pipeline to scaling existing systems for production.

**Q: Do you work remotely?**
A: Yes. Tensorhead is remote-first and works with clients worldwide. Based in Gothenburg, Sweden, but available for on-site work in Sweden and Europe when needed.

**Q: What size companies do you typically work with?**
A: Primarily startups and scaleups — companies that need senior AI/ML expertise but may not have a full in-house team yet. Also work with larger companies on specific projects or advisory engagements.

**Q: How long do typical engagements last?**
A: It varies — from a few weeks for advisory/strategy work to several months for end-to-end ML system builds. The discovery phase helps define the right scope.

**Q: Can you help evaluate AI tools and vendors?**
A: Yes. AI Strategy & Advisory services include vendor evaluation, build-vs-buy analysis, and technical due diligence — helping CTOs and product teams make informed decisions.

**Q: What industries do you work in?**
A: Industry-agnostic, with experience across fintech, e-commerce, healthtech, logistics, and SaaS. The core ML/AI skills transfer across domains.

**Q: Do you do fixed-price or hourly billing?**
A: Both, depending on project scope and clarity. Discovery and advisory work is typically hourly; well-scoped build projects can be fixed-price.

**Q: Can you help hire and build an internal AI team?**
A: Yes. Beyond hands-on engineering, Tensorhead can help define roles, interview candidates, and set up processes for an in-house AI/ML function.

---

## Svenska (Swedish)

### Om Tensorhead

Tensorhead AB är ett frilansande AI/ML-konsultbolag grundat av Patrik Larsson, en ingenjör med över 10 års erfarenhet inom maskininlärning och AI. Vi hjälper startups och scaleups att omvandla ambitiösa AI-idéer till produktionssystem som levererar verkligt affärsvärde.

**Plats:** Göteborg, Sverige
**Tillgänglighet:** Världsomspännande, remote-first

### Tjänster

- **ML-modellutveckling**: Skräddarsydda maskininlärningsmodeller från prototyp till produktion.
- **MLOps & Infrastruktur**: Skalbara ML-pipelines, modellserving, övervakning och CI/CD.
- **LLM & GenAI-lösningar**: RAG-system, finjustering, promptteknik och agentarbetsflöden.
- **Dataingenjörstjänster**: Datapipelines, feature stores och ramverk för datakvalitet.
- **AI-strategi & Rådgivning**: Teknisk due diligence, roadmap-planering och vägledning för AI-investeringar.

### Vanliga frågor (Svenska)

**F: Vilka typer av AI-projekt hanterar Tensorhead?**
S: Allt från skräddarsydd ML-modellutveckling och MLOps-infrastruktur till GenAI/LLM-lösningar och dataingenjörstjänster.

**F: Arbetar ni på distans?**
S: Ja. Tensorhead arbetar remote-first med kunder världen över. Baserade i Göteborg, men tillgängliga för arbete på plats i Sverige och Europa vid behov.

**F: Vilka företag arbetar ni vanligtvis med?**
S: Främst startups och scaleups — företag som behöver senior AI/ML-expertis men kanske inte har ett komplett internt team ännu.

**F: Kan ni hjälpa till att utvärdera AI-verktyg och leverantörer?**
S: Ja. Våra AI-strategi- och rådgivningstjänster inkluderar leverantörsutvärdering, bygg-kontra-köp-analys och teknisk due diligence.

**F: Inom vilka branscher arbetar ni?**
S: Branschoberoende, med erfarenhet inom fintech, e-handel, hälsoteknik, logistik och SaaS.

### Svenska söktermer / Swedish search terms

AI-konsult, AI konsult Göteborg, AI konsult Sverige, maskininlärning, ML-ingenjör, MLOps-konsult, generativ AI, GenAI-rådgivning, LLM-lösningar, dataingenjör, datapipelines, AI-strategi, teknisk rådgivning, frilans AI-konsult, maskininlärning konsulttjänster, AI-rådgivning Sverige, MLOps Sverige, AI-utveckling Göteborg

## Contact

- **Website:** https://www.tensorhead.com
- **Email:** patrik@tensorhead.com
- **LinkedIn:** https://www.linkedin.com/in/p-larsson/
- **Location:** Stavhopparegatan 5A, 41659 Göteborg, Sweden
- **Company registration:** Org.nr 559438-7853, VAT SE559438785301