f
finetuningrag

Sebastian G.

@finetuningrag

AI Data Engineer, RAG Pipelines, PDF to Markdown, Local LLMs

Deutschland
Deutsch, Englisch
Einige Informationen werden in englischer Sprache angezeigt.
Über mich
**Unsloth • MLX • RAG • LoRA/QLoRA • Local LLMs** Privacy-First RAG & Local LLM Fine-Tuning Tired of generic chatbots? I build secure, offline AI systems for your internal data. Custom RAG: Converting messy PDFs into clean Markdown (Python) with AnythingLLM, Ollama & LM Studio. Model Fine-Tuning: Custom LoRA/QLoRA training via Unsloth for your exact domain. 100% GDPR-Compliant: Powered by high-end Apple Silicon (128GB RAM) to keep sensitive corporate data completely local. Backed by a strong background in instructional design and content structuring. Let’s talk!... Mehr lesen

Kompetenzen

f
finetuningrag
Sebastian G.
offline • 
Durchschnittliche Antwortzeit: 1 Stunde

Meine Dienstleistungen

Datenanalyse-Beratung
I will analyze your documents and create a rag preparation roadmap
KI-Modell-Feintuning
I will finetune a custom llm locally for strict gdpr compliance

Portfolio

Arbeitserfahrung

kita

kita

Freiberufler • 7 yrs 8 mos

Media Educator & Digital Workflow Specialist | Daycares

Nov 2025 - Present • 11 mos

Specialized in digitizing complex documentation workflows and managing large volumes of qualitative data. Unstructured Data Management: Responsible for comprehensive developmental documentation, requiring the structuring and synthesis of massive amounts of qualitative, unstructured text data. Digital Transformation: Integrated digital tools into daily operations to optimize internal processes and build foundational knowledge architectures. Domain Expertise: Provided the deep domain knowledge necessary to structure human-centric data for later algorithmic processing and AI assistance."

Founder & Lead AI Data Engineer

Jan 2020 - Present • 6 yrs 9 mos

Spearheading the development of domain-specific AI solutions and local LLM architectures. AI & Model Fine-Tuning: Developed and fine-tuned the 'EleMo' model family (based on Mistral and Qwen architectures) using parameter-efficient techniques (LoRA, QLoRA) for specialized text generation. RAG Data Pipelines: Architected automated Python-based ETL pipelines to ingest, clean, and transform unstructured documents (complex PDFs) into high-quality, token-optimized Markdown for Retrieval-Augmented Generation. Local Infrastructure: Deployed fully offline, privacy-compliant AI knowledge bases using tools like AnythingLLM, DEVONthink, and custom Python scripts on high-performance Apple Silicon hardware."

Landeshauptstadt_Stuttgart

IT Project Consultant & Software Rollout Specialist | Landeshauptstadt Stuttgart

Landeshauptstadt Stuttgart • Vollzeit

Oct 2014 - Jul 2019 • 4 yrs 9 mos

Led strategic IT and digitization initiatives for a major municipal administration. Software Implementation: Directed the rollout and process consulting for specialized enterprise software across municipal departments. Cross-Functional Coordination: Collaborated closely with IT specialists, legal counsel, and management to ensure strict data compliance and project security. Asset Management: Held core responsibility for asset management and financial safeguarding of seven-figure municipal provisions. Strategic Decision Making: Prepared foundational data and decision-making frameworks for executive management and political committees."