Skip to content
View Dev-next-gen's full-sized avatar
🌐
Focusing
🌐
Focusing
939 followers · 209 following

Achievements

Achievement: Quickdraw

Achievements

Achievement: Quickdraw

Block or report Dev-next-gen

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
Dev-next-gen/README.md

morse

Typing SVG 
whoami
> Leo Camus -- @Dev-next-gen
> Paris, France
> Self-taught. No degree. Full stack from silicon to inference.

I build AI systems end-to-end -- from GPU kernel tuning to autonomous multi-agent pipelines. I run 80B parameter models locally. I automate offensive security at scale. I don't wait for the right tools to exist.

Contribution Snake

Core Expertise 
AI Infrastructure & LLM:
  llama.cpp - Ollama - LangChain - Multi-agent orchestration
  Local inference -- Qwen3-Coder 80B + 14B - ctx 262K - ROCm

Offensive Security:
  Bug bounty automation - Pentest pipelines - CVSS analysis
  nuclei - subfinder - katana - httpx - dalfox - jwt_tool

GPU & Compute:
  AMD / NVIDIA multi-GPU - ROCm - OpenCL - Vulkan
  Kernel-level tuning - Inference optimization

Systems & Backend:
  Linux (Ubuntu / Kali) - Python - Node.js - Rust - API REST
  MySQL - PostgreSQL - Supabase - self-hosted infra

OSINT & Geopolitics:
  Open-source intelligence pipelines - GDELT - ACLED
  Real-time geospatial analysis - Sanctions mapping
Projects 
[ WIP ] Geopolitical OSINT Platform — open-source Palantir alternative
    => Replaces classified sources with 6 tiers of legally-exploitable open data
    => GDELT - ACLED - SIPRI - Sentinel Hub - Copernicus - NASA FIRMS - ADS-B Exchange - OpenSanctions
    => LLM pipeline: multi-source ingestion -> entity graph -> geostrategy briefs
    -> 30+ integrated sources - real-time - fully local
    -> github.com/Dev-next-gen/osint-platform

  [ PROD ] OpenClaw — Autonomous bug bounty pipeline
    => 4 specialized LLM agents: recon - scan - analysis - reporting
    => Qwen3-Coder 80B (recon/scan) + Qwen3-14B (orchestration/analysis)
    => Full pipeline: subfinder -> httpx -> katana -> nuclei -> CVSS -> HackerOne report
    => Tested on real targets - rewards up to $10,000 Critical
    -> 100% local - multi-GPU - autonomous
    -> github.com/Dev-next-gen/distributed-agent-runtime

  [ LIVE ] AI Orchestrator
    => LLM orchestration — 80B strategy + 14B execution + quality control loops
    => STRATEGY pattern - score threshold - auto retry
    -> github.com/Dev-next-gen/ai-orchestrator

  [ LIVE ] Multi-Agent Framework
    => Lightweight multi-agent framework — ACP protocol, task graph, role-based agents
    => Tool registry - parallel execution - local LLM backends
    -> github.com/Dev-next-gen/multi-agent-framework

  [ LIVE ] GPU Cluster Lab
    => AMD/NVIDIA GPU cluster infrastructure — ~300 GPU deployment experience
    => ROCm kernel tuning - flash attention - multi-node benchmarking
    -> github.com/Dev-next-gen/gpu-cluster-lab

  [ LIVE ] Local LLM Stack
    => Production-grade local LLM deployment — llama.cpp, Ollama, GGUF, ROCm AMD
    => 14B to 80B - zero cloud dependency - benchmarked configs
    -> github.com/Dev-next-gen/local-llm-stack
CTF Flag 
contact
  |-- email      -> leo.camus23@gmail.com
  |-- linkedin   -> linkedin.com/in/leo-camus-4bb480304
  +-- site       -> nextgen-labs.net

Open to freelance missions, research collabs, or projects that shouldn't exist yet.

"I don't wait for technologies. I build them faster."

Pinned Loading

  1. ai-orchestrator ai-orchestrator Public

    LLM orchestration layer — coordinate 14B + 80B models, quality control loops, strategy/execution split

    1

  2. distributed-agent-runtime distributed-agent-runtime Public

    Distributed autonomous agent runtime — multi-LLM orchestration, task delegation, parallel execution (OpenClaw-based)

    1

  3. gpu-cluster-lab gpu-cluster-lab Public

    AMD/NVIDIA GPU cluster infrastructure — ~300 GPU deployment, ROCm, kernel tuning, multi-node benchmarking

    1

  4. local-llm-stack local-llm-stack Public

    Production-grade local LLM deployment stack — llama.cpp, Ollama, GGUF/GGML, ROCm AMD, 14B to 80B models

    1

  5. multi-agent-framework multi-agent-framework Public

    Lightweight multi-agent framework with ACP protocol, role-based agents, tool registry and parallel execution

    1

  6. osint-platform osint-platform Public

    Open-source intelligence platform — Palantir Gotham alternative. 6-level source integration, ontology graph, real-time threat analysis.

    3