Hi, I'm Dino
AI Developer
I build agentic AI systems at BlueCallom in Zurich.
My story so far
I moved to Switzerland in 2018. I spent six years in landscaping - mowing lawns, cutting trees and eventually leading a gartening team. The entire time during my landscraping adventure I was teaching myself to code at evenings.
In early 2025 I joined BlueCallom AG as an AI developer, where I now build agentic AI systems that automate business workflows from sales to compliance.
I'm also finishing a BSc in Applied Artificial Intelligence at IU International University.
AI Developer
BlueCallom AG, Zurich
BSc Applied Artificial Intelligence
IU International University
Switzerland
Menzingen, ZG
My path
Feb 2025 — Present · Zurich, Switzerland
AI Developer
BlueCallom AG- Build and ship end-to-end agentic AI systems that automate business processes
- Designed and shipped an agentic B2B sales pipeline covering lead qualification, outreach, email management, offer/order/invoice generation, and calendar scheduling.
- Develop new platform features in Python, Django, and SQL
- Integrate LLM providers (OpenAI, Anthropic, Perplexity) into the platform's agent layer.
- Design and build RAG pipelines, using Weaviate in the cloud and FAISS for local retrieval
Jan 2020 — Jan 2025 · Baar, ZG, Switzerland
Gardener → Head of Garden Maintenance
R. Zurcher Hauswart-Service GmbH- Started as a gardener and later department head. Managed a team of 8 across 170+ property contracts
Jun 2018 — Dec 2019 · Zug, Switzerland
Landscaper
Gartenbau A. Hurlimann GmbH
- Commercial and residential landscaping
Feb 2015 — Feb 2018 · Zagreb, Croatia
Warehouse Worker → Warehouse Manager
SALESIANER Gruppe- Ran daily warehouse operations and inventory management
What I've built
Multi-Tenant RAG Pipeline System
Provider-agnostic agentic RAG framework. An LLM agent receives a query, builds an execution plan, then loops through tools — semantic search, SQL queries, dependency checks — until it assembles a complete answer. Multi-tenant by design.
LLM Cloud Deployment
Self-hosted Google Gemma 4 (E2B-it)streaming inference API on AWS, served with vLLM on GPU-backed ECS with scale-to-zero. Six Terraform modules: multi-AZ VPC, ALB + WAF, a FastAPI auth proxy, CloudFront + S3 chat UI, and CloudWatch dashboards with alerting.
ArXiv AI Research Trends
Unsupervised learning analysis of 181k+ AI papers from ArXiv (2024-2026). Tested 9 embedding × clustering combinations, validated clusters against ArXiv categories using ARI and NMI, and identified growth trends via linear regression. Cluster labels generated automatically using Claude API.
Say hello
Feel free to reach out if you have any questions or would like to get in touch.
dino.smuc@gmail.com
Location
Menzingen, Switzerland