# AlamedaDev

> AlamedaDev is a reviewed AI agency in Barcelona, Spain, listed in the AIAgencies.eu register and last assessed 2026-09-10.

Canonical page: https://www.aiagencies.eu/agency/alamedadev
Generated: 2026-09-21 · Curated by Gabor Kiss · No paid placements — inclusion, score, and rank are never sold.

## Facts
- Location: Barcelona, Spain
- Website: https://www.alamedadev.com
- Hourly rate: not published
- Team size: 2-9
- Founded: 2009
- Services: AI Development, Generative AI, AI Agents
- Industries: SaaS & B2B, Healthcare
- LinkedIn: https://www.linkedin.com/company/alameda-dev

## About
AlamedaDev is a Barcelona AI development studio building agentic AI, retrieval-augmented generation and document and audio intelligence systems.

## Documented clients
Documented engagements: Tao Group Hospitality, Meter, and BcnByBike. All 3 carry a source URL on file.

## Curator assessment
Reviewed 2026-09-10 by Gabor Kiss against published evidence.

The AI-heavy work is documented in technical detail but anonymized and early-stage: the call-analysis system names its training volumes and a fine-tuned speech-to-text model at 95%+ accuracy, then stops at a pilot with selected agents, and the multi-agent diagnostics engagement is labeled a proof of concept.

The named clients—Tao Group Hospitality, Meter, BcnByBike—sit on the software side of the portfolio, so the two halves do not overlap.

Boundary-setting is explicit where it matters: the thyroid product states that it interprets and contextualizes but does not diagnose, and that those limits are enforced in prompts and architecture, not only in the interface.

Best for teams that want an AI prototype built against clinical or compliance boundaries and validated before scale, not an operator for a system already in production.

### Key strengths
- Case studies publish data volumes and model choices—10,000 calls and roughly 500 hours of audio to fine-tune the speech model—so the delivery estimate is checkable before you commit
- Reported outcomes on the diagnostics work are specific: 23% shorter consultations, 18% better report consistency and 12% more complex cases identified earlier
- Google Cloud and Vertex AI partner with published RAG and vector-search work, so the stack is legible from the project pages

### Good to know
- The two AI-heaviest studies are anonymized—ask for a speakable reference on the call-analysis system before shortlisting
- Both flagship AI engagements stop at pilot or proof of concept; ask what reached production, who operates it and what it costs to run
- No team page names ML engineers and the registry lists 2-9 people—clarify who builds and how a second concurrent build would be staffed

---
Source: AIAgencies.eu — the curated register of European AI agencies. Rate provenance: "agency-confirmed" = disclosed via submission or claimed profile; "directory-listed" = Clutch or two agreeing B2B directories (lone unverified sources are never written). Full method: https://www.aiagencies.eu/methodology · Rate index: https://www.aiagencies.eu/rates