Skip to content
View 6sLOGAN78's full-sized avatar
🎯
Focusing
🎯
Focusing
13 followers · 21 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Highlights

  • Pro

Block or report 6sLOGAN78

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

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

Report abuse
6sLOGAN78/README.md

6sLOGAN78 Name Header

AI/ML SYSTEMS  •  HIGH-PERFORMANCE BACKENDS  •  CONCURRENT ARCHITECTURES

The Learning Curve on Substack GitHub followers

Profile • Telemetry • Languages • Systems • Writing • Stack • Connections

Profile

I bridge the gap between autonomous deep learning systems and high-performance backend pipelines.

I am an engineer specializing in AI/ML systems and backend infrastructure. My work focuses on training specialized transformer architectures, optimizing inference pipelines, and designing high-throughput concurrent backend systems.

  • ⚡ Focus Areas: Deep Learning Architecture, Distributed Systems, Low-Latency Networking
  • 🧠 Research & Dev: Custom Transformer Pipelines, C++ CUDA Accelerators, Go Microservices
  • 🛠️ Design Philosophy: Minimalist, High-Performance, Mathematically Rigorous
  • ✍️ Writing: Contributor at The Learning Curve, the tech magazine run by NJACK, IIT Patna

Telemetry Statistics

Language Distribution

6sLOGAN78 Most Used Languages Pie Chart

Selected Systems

System What it is Stack
devRAG Enterprise RAG with hybrid BM25 + dense retrieval, reranking, tenant isolation and RBAC Go Python
RedrobRanker CPU-optimized 11-stage candidate ranking engine: BM25 + FAISS, cross-encoder reranking, local LLM rationales Python
NetSure Network intrusion detection pipeline on CICIDS2017 with DVC, MLflow and Optuna-tuned XGBoost Python Docker
SiLabs EFR32/ESP32 ICU telemetry with a stacking meta neural network for triage C
protask Task management on a Go (Echo) REST API with PostgreSQL, Redis queues and a React SPA Go TypeScript
show-its-work Deterministic KPI diagnosis engine where the LLM never computes a number Python

Writing

I write for The Learning Curve, the Substack tech magazine run by NJACK, the computing club of IIT Patna.

Subscribe to The Learning Curve

Architecture & Domains

THE THEORETICAL

THE TACTICAL

THE ORCHESTRATION

Terminal Connections

Pinned Loading

  1. cancer_pred_model cancer_pred_model Public

    End-to-end chest CT scan cancer classification using VGG16 (TensorFlow/Keras). Features a DVC-orchestrated 4-stage pipeline (ingestion → base model → training → evaluation), MLflow experiment track…

    Python

  2. protask protask Public

    Production-ready task management boilerplate Go (Echo v4) REST API with PostgreSQL, Redis/Asynq queues, Clerk auth & New Relic APM, paired with a React SPA (Vite + Tailwind CSS v4). Type-safe contr…

    TypeScript

  3. fraud-detection-optimization fraud-detection-optimization Public

    End-to-End Fraud Detection Pipeline with LightGBM, Optuna, Simulated Annealing & MLflow

    Python

  4. NetSure NetSure Public

    Modular, production-ready NIDS pipeline on CICIDS2017 using DVC for orchestration, MLflow for experiment tracking, and Docker for containerized deployment. Features hyperparameter-optimized XGBoost…

    Jupyter Notebook

  5. devRAG devRAG Public

    Enterprise RAG (Retrieval-Augmented Generation) system featuring document chunking, hybrid search (BM25 + Dense Embeddings), vector search, reranking, tenant isolation, and RBAC. Built with Go and …

    Python