Skip to content
View thatwhocode's full-sized avatar
0 followers · 1 following

Block or report thatwhocode

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
thatwhocode/README.md

Hi there! 👋 I'm Stanislav Utkin

🚀 Backend Engineer | MLOps & AI Infrastructure | Python & C++

I am a driven software engineer with a strong focus on building scalable backend systems, robust AI/ML infrastructure, and secure microservices. My expertise spans Clean Architecture, Natural Language Processing (NLP), and deploying Local LLMs (RAG pipelines) into production. I am passionate about taking complex concepts from zero to production and solving real-world business challenges without compromising data privacy.

🧠 What I Do

  • AI Infrastructure: Building Zero-Leak Enterprise RAG systems, integrating Local LLMs (Ollama, Llama 3), and working with Vector Databases (Qdrant).
  • Backend Architecture: Designing asynchronous, high-load APIs using FastAPI, Celery, Redis, and PostgreSQL with a strict adherence to Repository/Service patterns and Dependency Injection.
  • MLOps & NLP: Fine-tuning Transformer models (XLM-RoBERTa), building NER pipelines, and packaging ML cores into containerized, production-ready microservices.

🚀 Flagship Projects

1. 🔐 Enterprise AI SaaS Backend (Hybrid RAG)

  • Description: A production-ready, fully local Retrieval-Augmented Generation (RAG) backend designed for B2B environments requiring strict data privacy. Features asynchronous PDF ingestion, dual-memory architecture (Qdrant + Postgres), background task offloading, and a secure JWT authentication flow with Token Blacklisting.
  • Tech Stack: FastAPI, Qdrant, PostgreSQL, Celery, Redis, Ollama, Docker Compose.
  • Link: http://github.com/thatwhocode/rag_saas

2. 🛡️ Law Enforcement NER Pipeline (Neuro_app)

  • Description: Developed and deployed a complete ML system for Named Entity Recognition (NER) based on XLM-RoBERTa for a government contractor. Fine-tuned the model to extract weapon characteristics from unstructured text with 97% accuracy. Designed as a fully on-premise, containerized microservice.
  • Tech Stack: Python, PyTorch, Hugging Face, FastAPI, Docker, CI/CD.
  • Link: https://github.com/thatwhocode/neuro_app

3. ⚡ High-Performance HTTP Server

  • Description: Built a custom HTTP server from scratch in C++ to deepen my understanding of low-level networking, memory management, and concurrent connection handling using native Linux Sockets API.
  • Tech Stack: C++, Linux API, POSIX Sockets.
  • Link: https://github.com/thatwhocode/http_server_cxx

🤝 Let's Connect

Pinned Loading

  1. rag_saas rag_saas Public

    Python

  2. neuro_app neuro_app Public

    Python 1

  3. http_server_cxx http_server_cxx Public

    C++ 1