avtak.dev

Leipzig, Germany · open to ML and AI roles

Daniel
Hassanzadeh

ML and AI engineer. I build things people use.ML I measure honestly.guardrails for AI agents.

Legal name: Hassan Hassanzadeh Aliabadi

Daniel Hassanzadeh
avtak learning German
scroll

Selected work

Four things I built, and what each one proves.

01Live

Sprachvogel

A German course from A1 to B1. I built it and I run it on my own.

I am learning German myself, so I built the course I wanted. An LLM pipeline writes the content. Rule-based scripts check it, and two model reviews audit it before it ships. It is free and has no ads.

3,213words
224graded stories
30grammar lessons
~1,100automated tests

ProvesI finish things and run them alone.

Flutter webSupabaseCloudflareClaude API
sprachvogel.com
02Open source

divan

Guardrails for AI coding agents, in code instead of prompts.

Agents skip rules written in a prompt, so I moved the important ones into code. A hook refuses destructive commands. A lint refuses a fact in two places or a secret in a commit.

ProvesI engineer agent workflows that hold up.

PythonGitMIT
GitHub
03Paper in preparation
Daniel presenting his depression detection research

Early depression detection

MSc research, University of Malaya.

A long-context transformer (Mental-Longformer, 4,096 tokens) reads a user's post history and looks for early signs of depression. It is measured on the unseen eRisk 2025 collection.

0.740F1
0.958AUC
909unseen users

ProvesI measure on data the model has never seen.

PyTorchTransformersGradioMCP
MCP agent demo
04Coursework

Health prediction models

Two classic machine learning projects from my studies.

Obesity risk prediction from lifestyle data, and breast-cancer survival classification. Both are on GitHub with their notebooks.

0.92F1, obesity risk
0.75ROC AUC, survival

ProvesI know the classic ML toolkit, not only LLMs.

scikit-learnpandas
Obesity   Breast cancer

About

Software engineer first, then data science, now Leipzig.

I started as a software engineer and moved into data science for my MSc. My research looked at early depression detection, and it took me to the Information Retrieval group at USI Lugano as a visiting researcher. I chose that topic because I have seen how hard it is to ask for help when you are far from home, and I want AI to help people get that help earlier.

In Malaysia I worked on product and AI: first as product lead and AI engineer for an event platform, then on automation and AI tools for a food retailer.

Now I live in Leipzig. I am learning German, and I build the tools I use to do it.

I grew up watching my father build PCs, and I still learn that way: watch someone good, then do it myself. Before tech, I ran my own coffee shop. I still think best with a coffee in hand.

Online I am avtak, an old gaming name I never gave up.

  1. 2025 to 2026
    Automation & AI Engineer
    Gulsanco, Malaysia
  2. 2025
    Visiting Researcher
    USI Lugano, Information Retrieval group
  3. 2024 to 2025
    Product & AI Engineer
    Exalumen Technologies, Malaysia
  4. 2023 to 2025
    MSc Data Science, Distinction
    University of Malaya
  5. 2019 to 2023
    BSc (Hons) Software Engineering
    Asia Pacific University and De Montfort University

Contact

Let us build something useful.

I am open to ML, AI and software roles in Leipzig and across Germany.

hello@avtak.dev