anbat3 anbat3

Hi, I’m @anbat3!

I'm a computational linguist and editor specialized in NLP, Generative AI (LLMs) and conversational interfaces.

Currently, I’m designing intelligent filtering systems using multi-LLM architectures for energy market analysis.

I’m passionate about all things language — from syntax trees and linguistic theory to the algorithms that try to mimic it on LLMs.
Before stepping into tech, I was a book editor — a role that sharpened my attention to detail and my appreciation for the mechanics of language.
I believe that throughout life you only get a handful of chances to make bold choices — and when they come, you have to take them.
Applying that last motto, I am currently enrolled in an internships as a Quant Developer.
I am also looking for the next place to put everything I know into practice.

Programming & Data: Python (pandas, numpy, regex, SQL), web scraping, API integration, data visualization (Matplotlib, Seaborn, Plotly). XML/JSON parsing, webscraping, regex-based normalization, annotation workflows, and corpus linguistics. Strong understanding of control flow, data structures, and modular design.
MLP & AI: LLM optimization, RAG, recursive prompt engineering, spaCy, NLTK, Hugging Face Transformers, POS tagging, NER, sentiment analysis (TextBlob, VADER), embeddings, ASR/TTS.
LLMs & Generative AI: Prompt engineering (basic and recursive), fine-tuning concepts, evaluation with MAE, standard deviation, semantic coherence. Tools: LangChain, Hugging Face Transformers, OpenAI API, LlamaIndex.
Conversational AI: Design and development of virtual assistants using Rasa and DialogFlow. Integration of TTS and ASR modules. Dialog flows, user-centered design, accessibility, prototyping (Figma, Miro).
Visualization & Dashboards: Data storytelling with Matplotlib, Seaborn, Plotly. Dashboard creation with Streamlit for NLP applications and analysis tools.
Dev Tools & Cloud: Git/GitHub, VS Code, Streamlit, familiarity with AWS, AI coding assistants (Cursor, Windsurf).
Editing & Design: HTML, CSS, XML, InDesign, Illustrator, Photoshop, Genially, Procreate.
Languages & Communication: Native Spanish, fluent English (C1+). Comfortable writing technical documentation and working in cross-functional teams.

DEL.IA – Multilingual Conversational Assistant for Hate Speech Detection:
- Developed in Spanish, English, and French using Rasa + Hugging Face Transformers.
- Integrated data pipelines for real-time hate speech classification and multilingual understanding.
Internship Project – AI-driven News Relevance Evaluation for Investment:
- Multi-model LLM pipeline with recursive prompt engineering for financial news analysis.
- Semantic classification, NER, sentiment analysis, knowledge graph construction.
- Interactive dashboards for performance evaluation and investment insights.