Applied AI Vienna

Applied AI systems for complex workflows.

We build custom AI systems, guide AI strategy and implementation, and teach AI literacy — from prototypes to operational systems and funded innovation projects.

Discuss a project View capabilities

ePATH Live · 4 countries

Focus Custom AI systems · agents · automation

Austrian company Custom AI systems AI agents & automation Strategy & implementation AI literacy & education

{{/* Oversized faint hex mark, partially clipped by .about's overflow:hidden. Strokes self-draw via stroke-dasharray on viewport entry (.in via the shared .reveal IntersectionObserver). Path coordinates lifted from assets/logo/logo_dark.svg's #g44 group; pathLength="1" normalizes lengths so a single dasharray:1 covers every path. */}}

01 · About

A young Austrian AI company built for implementation.

synapsyx GmbH is a Vienna-based AI company working across development, strategy, and education. We work mainly with education, research, social-sector, and publicly funded organizations across Austria and the DACH region.

2025 Founded · Vienna, AT

We build custom AI systems for workflows where generic tools fall short: internal knowledge work, document-heavy processes, education platforms, research and innovation projects, and operational decision support.

Our work combines hands-on AI experimentation, software development, context engineering, structured evaluation, and project delivery. We combine strategy, implementation, and education — from identifying suitable use cases to building systems, evaluating them, supporting adoption, and gradually turning proven components into reusable products.

Founded

2025

Registered Austrian GmbH, based in Vienna.

Focus

Applied AI

Custom systems, agents, internal tools, platforms, and automation workflows.

Delivery

In production

Our systems reach production — ePATH runs live for students in four countries within an Erasmus+ project.

Operations

Internal tooling

Structured AI workflows for research, documentation, project development, and quality review.

01 / Applied

Built around real workflows.

We build around real workflows, not generic AI demos. The goal is practical usefulness, reviewability, and fit with existing organizational processes.

02 / Human-reviewable

Review points kept explicit.

Our systems are designed with explicit human review points, structured outputs, source awareness, and clear responsibility.

03 / Implementation-focused

Strategy through to adoption.

We combine strategy, technical prototyping, software development, evaluation, and adoption support in one small team.

02 · Capabilities

What we do.

We work where AI has to fit into real processes, sensitive information flows, and practical implementation constraints — across development, strategy, and education.

03 Pillars

01 / Develop

AI solution development

Custom AI tools, platforms, pipelines, internal applications, agents, and workflow automation for organizations that need more than an off-the-shelf chatbot.

Custom AI applications and platforms
AI agents and workflow automation
Internal tools and dashboards
API and platform integrations
Prototype-to-implementation development

02 / Implement

AI strategy and implementation

We help teams identify where AI is useful and feasible, then deliver it — including R&D and funded innovation projects, with evaluation built in.

AI readiness and use-case selection
Feasibility and risk mapping
Implementation roadmaps
R&D and technical work packages
Evaluation, quality control, and funded-project support

03 / Educate

AI literacy and education

We build AI capability inside teams and institutions through hands-on, practical formats grounded in real implementation experience.

AI-literacy workshops and guest teaching
Prompt- and context-engineering training
AI-supported project-management formats
Adoption and enablement support

03 · Our work

Selected systems and work.

Many AI projects involve sensitive workflows, data, or strategic plans, so we show selected public examples, anonymized project snapshots, and internal systems we use ourselves. More detail is available where appropriate.

05 Selected systems

{{ $epathPoster := "" }} {{ $epathWebm := "" }} {{ $epathMp4 := "" }} {{ with $epath }} {{ with .Params.video }} {{ with resources.Get .poster }}{{ $epathPoster = . | fingerprint }}{{ end }} {{ with resources.Get .webm }}{{ $epathWebm = . }}{{ end }} {{ with resources.Get .mp4 }}{{ $epathMp4 = . }}{{ end }} {{ end }} {{ end }} {{ if and $epathWebm $epathMp4 }}

Request a deployment

P-01 In production · 4 countries

ePATH

AI-supported career orientation and entrepreneurship reflection

ePATH is an AI-supported tool for career orientation and entrepreneurship reflection. It combines structured self-assessment logic with an AI counselor experience to help students explore career direction. ePATH is live in production within the Green Edu Seeds Erasmus+ project, used by students in Vietnam, Thailand, Germany, and Italy — embedded into MoodleCloud via LTI 1.3. It supports reflection and orientation; it does not make admissions, employment, or other high-stakes decisions.

Hybrid design combining skippable modular questionnaires with open-ended AI counselor dialogue
Questionnaire and AI recommendations benchmarked against O*NET occupational data during development
Embedded directly into university LMS environments via LTI 1.3

Internal tool · operational workflow

Internal AI workflows for project development

We use structured AI workflows internally to support research, project development, documentation, quality review, and implementation planning. These systems help us structure complex information, compare assumptions, prepare implementation plans, and keep human review explicit.

Research and source organization
Requirements and work-package structuring
Documentation support
Review and quality-control steps
Human responsibility kept explicit

Ask about our approach

Internal R&D · under development

R&D discovery and feasibility workflows

Internal workflows for exploring problem spaces, comparing solution paths, generating technical hypotheses, and identifying feasible AI project concepts.

Problem-space exploration
Solution-path comparison
Technical hypothesis generation
Feasibility and implementation assessment
Early-stage R&D structuring

Internal tooling · project capability

Agentic research and drafting workflows

Controlled agent workflows for repetitive research, classification, drafting, comparison, and information-management tasks — designed with review points rather than blind automation.

Multi-step research pipelines
Classification and comparison workflows
Structured drafting support
Reviewable intermediate outputs
Human-in-the-loop control

Prototype · capability demo

Knowledge and document workflow prototypes

Prototype workflows combining document processing, retrieval, structured outputs, and human review for domain-specific decision support.

Document processing
Retrieval and structured context
Domain-specific output formats
Review and correction loops
Decision-support interfaces

04 · Approach

Fast experimentation, disciplined implementation.

AI projects move quickly, but serious implementation still needs structure. We work in short cycles that make assumptions visible, test technical feasibility early, and keep review points explicit. We keep people in control of meaningful decisions, design for privacy and accessibility, and avoid training proprietary models.

05 Steps

Step 01

Understand

We map the workflow, users, constraints, available data, risks, and success criteria before choosing the technical path.
Step 02

Prototype

We build quickly to test assumptions, validate technical feasibility, and make the solution tangible.
Step 03

Implement

We develop pipelines, agents, interfaces, integrations, and internal tools that fit the real workflow.
Step 04

Evaluate

We test quality, reliability, usability, source handling, and business relevance before scaling the system.
Step 05

Scale

We improve, document, transfer knowledge, and support adoption so the system can become part of everyday work.

05 · Founders

Founder-led and implementation-focused.

synapsyx is led by its two founders, who combine AI systems development, product work, and AI strategy with backgrounds in economics, law, and organization-building. We operate as a small core team and bring in specialist collaborators and partners on individual projects.

02 Vienna

{{ range $i, $slug := $teamSlugOrder }} {{ $m := index $teamPagesBySlug $slug }} {{ if $m }} {{ $idx := printf "%02d" (add $i 1) }} {{ $copy := index $teamCopy $slug }} {{ $name := index $copy "title" | default $m.Params.name }} {{ $role := index $copy "role" | default $m.Params.position }} {{ $bio := index $copy "bio" | default $m.Params.bioShort }} {{ $photo := "" }} {{ with $m.Params.image }} {{ $photo = resources.Get . | fingerprint }} {{ end }} {{ $initials := "SX" }} {{ with $name }} {{ $parts := split . " " }} {{ $a := substr (index $parts 0) 0 1 }} {{ $b := "" }} {{ if gt (len $parts) 1 }}{{ $b = substr (index $parts 1) 0 1 }}{{ end }} {{ $initials = printf "%s%s" $a $b }} {{ end }}

{{ if $photo }} {{/* Portrait renders in an 84×84 circle. Serve 1x at native and 2x for retina via srcset so non-retina viewers don't download a 168×168 sized for high-DPR displays. */}} {{ $cropped1x := $photo.Fill "84x84" }} {{ $cropped2x := $photo.Fill "168x168" }} Portrait of {{ $name }}

06 · Partners

We work inside EU programmes, education platforms, and implementation networks.

Our work is connected to education, innovation, and project-development contexts across Europe and Southeast Asia. Partner logos are shown for the relationships and deployment contexts where public reference is appropriate.

{{ printf "%02d" (len .Site.Data.partners) }} Active collaborations

{{ range .Site.Data.partners }} {{ $pname := .name }} {{ $pdesc := .description }} {{ $purl := .url }} {{ $logo := resources.Get .logo }} {{ with $logo }} {{/* Partner logos render in a 60×60 glyph; ship a fitted raster (max 120px for retina) so a 1120×947 source doesn't send ~320KB to a 60-pixel slot. SVGs pass through unmodified. */}} {{ if ne .MediaType.SubType "svg+xml" }} {{ $logo = .Fit "120x120" }} {{ end }} {{ $logo = $logo | fingerprint }} {{ end }} {{ $initials := "" }} {{ with $pname }} {{ $parts := split . " " }} {{ $a := substr (index $parts 0) 0 1 }} {{ $b := "" }} {{ if gt (len $parts) 1 }}{{ $b = substr (index $parts 1) 0 1 }}{{ end }} {{ $initials = upper (printf "%s%s" $a $b) }} {{ end }}

07 · Footprint

From Vienna, into live AI workflows and education platforms.

synapsyx builds from Austria and collaborates across European and international project contexts. Our current footprint reflects where our team, partners, and deployed education work are active.

{{ $depCount := len (index .Site.Data.deployments "deployments") }} {{ $satCount := 0 }} {{ with index .Site.Data.deployments.hq "satellites" }}{{ $satCount = len . }}{{ end }} {{ printf "%02d" (add (add 1 $depCount) $satCount) }} Locations · 2026

{{/* Globe mounts into .footprint-canvas via globe.js once the section enters the viewport. The .footprint-list below carries the same data as text so the section reads correctly without JS, with screen readers, and in print. data-landmask-url points at the precomputed Stripe-style dot- matrix landmass dataset (regenerated by scripts/generate-landmask.py). globe.js fetches it lazily once the canvas mounts. */}} {{ $landmask := resources.Get "v2/landmask.json" }} {{ if hugo.IsProduction }}{{ $landmask = $landmask | resources.Fingerprint }}{{ end }}

{{ .city }} · {{ .region }}

{{ .role }}

{{ .description }}
{{ range .satellites }} {{ $pinIdx = add $pinIdx 1 }}

{{ .city }} · {{ .region }}

{{ .role }}

{{ .description }}

{{ end }}
{{ .region }}

{{ .role }}

{{ .description }}

08 · Work with us

Work with us.

We are selectively expanding our team and collaborator network for applied AI systems, automation workflows, agents, and R&D implementation.

We are interested in people who can think clearly, build quickly, and work responsibly with AI in real organizational contexts. We are especially open to AI systems builders, automation engineers, product-minded researchers, and collaborators with domain or implementation expertise.

Introduce yourself

AI systems builders
Automation engineers
Product-minded researchers
Domain & implementation collaborators

Exploring an AI project, prototype, or implementation challenge?

Discuss a project

or email {{ $contactEmail }}