---
title: "Alexandre Donciu-Julin (@alexdjulin) | Hashnode"
description: "AI Software Engineer | Technical Director | Python Allrounder. Read the latest articles by Alexandre Donciu-Julin on Hashnode."
url: "https://hashnode.com/@alexdjulin"
---

@alexdjulin

# Alexandre Donciu-Julin

@alexdjulin · Berlin, Germany · Joined March 2021

AI Software Engineer | Technical Director | Python Allrounder

### [Training and deploying a chat-moderator using PyTorch and AWS SageMaker](https://dev.alexdjulin.ovh/training-and-deploying-a-chat-moderator-using-pytorch-and-aws-sagemaker?utm_source=hashnode&utm_medium=feed)

A simple BERT-based model for toxic comment classification, implemented in PyTorch, trained and deployed on SageMaker

11 min read · [dev.alexdjulin.ovh](https://dev.alexdjulin.ovh/training-and-deploying-a-chat-moderator-using-pytorch-and-aws-sagemaker?utm_source=hashnode&utm_medium=feed)

[Mar 14, 2025](https://hashnode.com/posts/training-and-deploying-a-chat-moderator-using-pytorch-and-aws-sagemaker/67d3f170eb668286f5a6ac27)

### [Mistral 7B - InDepth Paper Presentation](https://dev.alexdjulin.ovh/mistral-7b-indepth-paper-presentation?utm_source=hashnode&utm_medium=feed)

What makes Mistral 7B so efficient, for such a small model?

1 min read · [dev.alexdjulin.ovh](https://dev.alexdjulin.ovh/mistral-7b-indepth-paper-presentation?utm_source=hashnode&utm_medium=feed)

[Dec 20, 2024](https://hashnode.com/posts/mistral-7b-indepth-paper-presentation/6765534ad410e66635885710)

### [Bringing Alice to Life in 7 Days](https://dev.alexdjulin.ovh/bringing-alice-to-life-in-7-days?utm_source=hashnode&utm_medium=feed)

Your own personal AI Librarian powered by Langchain & RAG

1 min read · [dev.alexdjulin.ovh](https://dev.alexdjulin.ovh/bringing-alice-to-life-in-7-days?utm_source=hashnode&utm_medium=feed)

[Dec 18, 2024](https://hashnode.com/posts/bringing-alice-to-life-in-7-days/6762eb800be88bc1ef59c660)

### [Chatting with my movie-advisor using a langchain agent, RAG tools, Xata PostreSQL and TMDB](https://dev.alexdjulin.ovh/chatting-with-my-movie-advisor-using-a-langchain-agent-rag-tools-xata-postresql-and-tmdb?utm_source=hashnode&utm_medium=feed)

https://youtu.be/\_nmiwgwRHxM Github repo This is a demo of my movie-advisor project: I created a chatbot driven by a langchain agent and RAG tools. The agent is handling all conversations between me and the LLM. Based on my queries, it will search ...

1 min read · [dev.alexdjulin.ovh](https://dev.alexdjulin.ovh/chatting-with-my-movie-advisor-using-a-langchain-agent-rag-tools-xata-postresql-and-tmdb?utm_source=hashnode&utm_medium=feed)

[Aug 10, 2024](https://hashnode.com/posts/chatting-with-my-movie-advisor-using-a-langchain-agent-rag-tools-xata-postresql-and-tmdb/66b7a9c10ea9fe185590b708)

### [Create a chatbot Intent Classifier based on fine-tuned GPT3](https://dev.alexdjulin.ovh/intent-classifier-based-on-finetuned-gpt3?utm_source=hashnode&utm_medium=feed)

Intent classification is a pivotal process in natural language processing (NLP) where an AI system identifies the purpose or intention behind a user's input. This task involves analyzing a user's message and categorizing it into predefined intents, s...

2 min read · [dev.alexdjulin.ovh](https://dev.alexdjulin.ovh/intent-classifier-based-on-finetuned-gpt3?utm_source=hashnode&utm_medium=feed)

[Jun 10, 2024](https://hashnode.com/posts/intent-classifier-based-on-finetuned-gpt3/6666f74d95e7577620a8bf04)
