SYSTEM ONLINE UTC --:--:-- WESSELING / DE-NW NOW BUILDING — HEALTHCARE-AGENT-V3 BOOKING Q3 / 2026 REPLY < 24H
v.2026.04 / booking Q3 · independent consultant

AI agents that actually run in production.

Independent AI & IT consulting from Germany. I turn real workflows into reliable agent systems — built end-to-end, instrumented, and EU-compliant by design.

Start a build View selected work
EU AI Act GDPR Made in Germany
Aidin Abbasi - AI Consultant Germany
/ profile_ Aidin Abbasi AI & IT Consultant · Wesseling, DE
01 — capabilities
i.

End-to-end delivery from use case to running system.

I work end to end: define use case and user journey, design prompts and agents, integrate tools and data, ship a reliable v1, then improve based on real usage instead of theory.

01.
Agent systems
Goal-driven agents with memory, tool use, guardrails, retries, and human handoff. Built with LangChain, CrewAI, and custom orchestration.
LangChainCrewAI
02.
Full-stack delivery
Python, React, FastAPI, Supabase, strong SQL, vector DB RAG, n8n orchestration, Azure & AWS deployments.
PythonReactAzure
03.
Hyper-automation
Assistants across Slack, Telegram, email, voice and web — alerts, summaries, lead handling, multi-channel ingestion.
SlackTelegramVoice
04.
Retrieval & grounding
Focused sources, embeddings, ranking — answers stay accurate, traceable and useful. EU-hosted infrastructure for sovereignty.
RAGpgvector
05.
LLM operations
Structured outputs, evaluation checks, monitoring, cost and rate control. Drift detection and prompt versioning baked in.
EvalsLLMOps
06.
Sovereign AI
EU AI Act, GDPR and NIS2 profiling, draft compliance reports, EU-hosted infra in Frankfurt, multi-tenant isolation.
EU AI ActGDPR
02 — selected_work
ii.

Production-grade work across healthcare, EU compliance & private knowledge.

Recent builds — each shipped, instrumented, and improved based on real usage. Architecture diagrams & demos available on request.

/01
AI Digital Twin Health Coach screenshot
Healthcare AI · Conversational

AI Digital Twin — Personal Health Coach

Persona-based health coach turning lab results, sleep, recovery and activity into clear, grounded conversations. Multi-user demo with live metric panels and safe-by-default guardrails.

LLMRAGHealthcareGuardrails
4 data sources unified
/02
Azure Health AI architecture document screenshot
Sovereign Cloud · Production

Azure Health AI Platform

End-to-end production architecture. 4 ingestion channels, pseudonymized storage in Frankfurt, n8n orchestration, full EU data residency, observability built in from day one.

Azuren8nEU AI ActFrankfurt
EU data residency
/03
ComplyAI EU AI Act compliance copilot screenshot
RegTech · Multi-Tenant SaaS

ComplyAI — EU AI Act Copilot

Multi-tenant SaaS profiling AI systems against EU AI Act, GDPR and NIS2. Document upload, AI-driven profiling, downloadable draft compliance reports — strict per-company isolation.

EU AI ActGDPRNIS2Multi-tenant
3 regulations covered
/04
Interview Vault private knowledge platform screenshot
Full-stack · Private Knowledge

Interview Vault

Private React SPA with Supabase auth, role-based admin routes, signed-URL uploads and n8n onboarding webhooks. Working blueprint for invite-only community apps.

ReactSupabaseRBACWebhooks
RBAC by design
~/builds/log.txt — append-only
14:23:08SHIPPEDCV builder — personalizes resumes at scale[200 OK]
14:18:42SHIPPEDDaily AI news agent[200 OK]
14:11:19SHIPPEDInsurance claim triage assistant[200 OK]
13:58:33SHIPPEDMulti-model routing gateway[200 OK]
13:42:07SHIPPEDInstagram DM keyword assistant[200 OK]
13:21:55SHIPPEDVoice-enabled sales copilot[200 OK]
12:54:18SHIPPED500+ source analytics dashboard[200 OK]
03 — who_i_help
iii.

For founders & operators who want AI that delivers measurable value.

The people I work with

Founders, managers, product/ops teams, and technical teams looking to deploy AI that delivers measurable value.

  • Founders shipping their first AI product
  • Product & ops leads automating real workflows
  • Technical teams needing senior AI delivery
  • Corporate innovators inside German SMEs

Typical use cases

Most engagements fall into a handful of repeatable shapes — well-understood problems that benefit from senior delivery.

  • Internal AI assistants grounded on docs and SOPs
  • Slack & Telegram assistants for alerts and summaries
  • Lead intake and triage automation
  • Support copilots with reliable retrieval (RAG)
  • Automating repetitive operations workflows
04 — process
iv.

A simple, repeatable path from idea to running system.

Step 01

Define the outcome

Week 1

Align on a measurable goal — cost savings, revenue lift, or a compliance milestone. Map current workflow, identify bottlenecks, agree on success metrics before a single line of code.

Step 02

Rapid proof-of-value

Week 2 – 5

Deliver a production-ready MVP within weeks, not months. Real data, real users, real feedback loop — built on infrastructure that can scale rather than throwaway demo code.

Step 03

Launch & monitor

Week 6 – 8

Ship to production with proper observability — usage analytics, KPI dashboards, AI behaviour tracking, drift detection, and cost monitoring. You can see what the system is actually doing.

Step 04

Iterate & scale

Month 3+

Optimise based on real usage signals, then prepare for rollout across departments. New use cases plug into the same foundation rather than starting from scratch each time.

Step 05

Upskill the team

Ongoing

Documentation, runbooks, and hands-on training so your team owns the system long after I'm gone. No vendor lock-in — the goal is autonomy, not dependency.

EU AI Act
GDPR
Data in DE
Privacy by Design
Made in DE
05 — contact
v.

Ready to build AI that actually works?

Tell me what you're building. If it's a fit, I'll send back a short discovery-call invite — usually within one business day.

I reply within one business day

Your data stays in Germany — no tracking

EU AI Act & GDPR compliant by design

Free 30-minute discovery call