Skip to content
View DioBr4nd0's full-sized avatar
πŸ’­
πŸ¦€πŸ¦€πŸ¦€
πŸ’­
πŸ¦€πŸ¦€πŸ¦€

Block or report DioBr4nd0

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 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
DioBr4nd0/README.md

πŸ‘‹ Good to have you here!!

🌱 About myself

Rupesh Prajapati

Systems Engineer (AI & Crypto) | IIT Roorkee Alumnus

I architect and build high-performance systems, infrastructure, autonomous agent state machines, and low-level systems. I specialize in moving AI out of fragile API wrappers and into robust, scalable, and hardware-accelerated production environments.

🧠 Core Competencies

  • AI Infrastructure: High-throughput LLM Inference, GPU/CUDA Acceleration, RAG Architecture, Evals (LangSmith, Langfuse).
  • Crypto: L2s, Ethereum, Rollups, Data availability solutions, Smart Contracts.
  • Agentic Systems: LangGraph, Multi-Agent Orchestration, n8n Automation Workflows, Local LLMs (Ollama).
  • High-Performance Systems: Rust, Distributed Databases, Tokio, low latency, High performance computing, eBPF.
  • Production Engineering: Kubernetes, Helm, CI/CD Pipelines, Distributed Systems.

⚑ Systems Architecture & Engineering

1. High-Throughput LLM Inference Systems (Rust, CUDA, HPC)

  • Architected scalable backend systems natively in Rust for managing and routing high-volume LLM requests in production environments.
  • Engineered core infrastructure components including low-latency request batching, rigorous rate limiting, and comprehensive observability pipelines.
  • Integrated hardware-accelerated execution layers utilizing CUDA, completely bypassing standard high-latency API bottlenecks to achieve bare-metal inference performance.

2. Cero (L2 Distributed Rollup Sequencer)

  • Worked on transaction sequencing/ordering, Validator selection, Rewards Distrubuion.
  • Block Propagation with CometBFT, Geth l1 settlement, DA solutions (Celestia, Avail)
  • Polygon/Erigon CDK integration, Distributed database sync, p2p

3. Autonomous Agentic Systems & State Machines (LangGraph, Ollama, n8n)

  • Architected autonomous, multi-agent systems using LangGraph and local LLM inference (Ollama) to execute complex, multi-step cognitive workflows.
  • Engineered CLI-based agentic tools featuring graph-based state machines capable of iteratively writing, executing, testing, and self-correcting code dynamically.
  • Designed and deployed human-in-the-loop (HITL) automation pipelines utilizing n8n for high-signal data scraping, dynamic context assembly, and reliable workflow execution.

4. Bare-Metal & Network Security Systems (Rust, eBPF, CPU)

  • Architected highly performant, bare-metal network systems utilizing Rust and eBPF for secure, kernel-level packet parsing and manipulation.
  • Engineered decentralized, high-throughput execution environments inspired by advanced blockchain infrastructure to ensure robust, fault-tolerant system design with zero single points of failure.

Linkedin Badge

Gmail Badge


πŸ“Ά Stats:


GitHub Streak

Pinned Loading

  1. iroh-libp2p iroh-libp2p Public

    Rust 1 1

  2. chip8-emulator chip8-emulator Public

    Rust 1

  3. go-microservice go-microservice Public

    Go

  4. rust-interpreter rust-interpreter Public

    Rust

  5. rust_port_sniffer rust_port_sniffer Public

    Rust