Skip to content
View joshconkel's full-sized avatar
  • Docusign
  • Rio Vista, CA / Oakland, CA
  • 13:09 (UTC -07:00)
  • LinkedIn in/josh-conkel

Block or report joshconkel

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
joshconkel/README.md

Josh Conkel

Lead Product Security Engineer  |  Docusign  |  Rio Vista / Oakland, CA

LinkedIn Email GitHub


What I Do

Product security engineer with 20+ years building security programs that engineering teams actually adopt. I work at the intersection of secure SDLC, agentic AI security, and software supply chain, translating risk into practical controls that ship into CI/CD pipelines without slowing delivery.

At Docusign I lead our Security Champions program, build security tools for developers to use on a daily basis, and coding security guardrails across product teams. Before that: principal-level consulting at KPMG and EY (AppSec, data security, and regulatory readiness for multiple large clients), and a DevSecOps practice lead at Slalom.

Currently building open tooling for the MCP and agentic AI attack surface, a risk area that is moving faster than most teams' defenses.


Currently Building

mcp-sentinel  active

Static and dynamic security auditor for MCP (Model Context Protocol) servers - the tool layer between LLMs and the systems they act on. Maps findings to OWASP MCP Top 10, OWASP Top 10 for Agentic Applications (2026), and MITRE ATLAS via a pluggable, versioned rule engine.

Built because the agentic AI attack surface (tool poisoning, supply chain compromise, over-permissioned schemas, prompt injection via tool results) is largely undefended and moving into production at speed.

python   mcp   llm-security   agentic-ai   appsec   owasp


Veritas-POC  active

An AI driven local code security scanner built to help developers perform more effective manual secure code reviews. This approach an agentic AI approach with JSON artifact pipelines to let developers know where the needles are in the haystack.

The pipeline scans a directory of source files, runs each file through a chain of specialized security agents (scope → threat model → hypotheses → evidence → fix → gate), and produces structured JSON artifacts and a Markdown report per file, plus a merged summary across all files.

python   agenticAI   security-champions   securecode   llm   owasp


Focus Areas

Program & Process

  • Security Champions Programs
  • Secure SDLC (Shift-Left)
  • Threat Modeling & Secure Design Reviews
  • Vulnerability Management Automation
  • Third-Party / OSS Governance
  • Regulatory Readiness (FedRAMP, FISMA, ISO 27001, SOX)

Engineering & Tooling

  • CI/CD Security Guardrails (policy-as-code, secrets, scanning)
  • Software Supply Chain (SLSA, code signing, SBOM, integrity)
  • Cloud Security (AWS, Azure)
  • Secrets Management
  • Data Security & DLP
  • Application Security Architecture

Emerging (2025-2026)

  • MCP Server Security
  • Agentic AI Threat Modeling
  • LLM Security (prompt injection, tool misuse)
  • AI Supply Chain Risk
  • AI-Assisted Security Automation

Featured Projects

Project What It Is Stack
mcp-sentinel MCP server security auditor (OWASP MCP Top 10, Agentic Top 10, MITRE ATLAS) with pluggable multi-source rule engine Python
security-champions-kit 36-month Security Champions program framework (intake, enablement tracks, metrics, governance playbooks) built from real program experience Markdown / Templates
Veritas-POC Multi-stage agentic LLM pipeline for automated security code review Python
Dome9toDD Dome9 cloud security findings bridge to DefectDojo for centralized vulnerability tracking Python

Certifications

CISSP OSCP AWS SAA AWS ML DevSecOps CISM

The CISSP (2007) paired with OSCP (2024) along with other certifications along the way reflects both program-level governance experience and continued investment in hands-on technical depth.


Tech

Python Java Bash AWS Azure Terraform Docker Kubernetes GitHub GitLab


Education

M.S. Engineering Management - Ohio University  |  B.S. Computer Engineering - Shawnee State University


Selected Wins (From the Field)

  • 80%+ reduction in security review time (multiple sprints to a single sprint before release) by delivering end-to-end pipeline automation of application security scanning at Slalom (global diagnostics client)
  • Data Security Program Office for one of the largest U.S. utilities, safeguarding data for 40M+ customers and 50K+ employees across cloud and on-prem environments (KPMG)
  • Software Supply Chain Security program at Docusign: code signing, artifact integrity verification, and dependency risk management across multiple products
  • Security Champions program at Docusign: gamified hands-on training (Security Journey), self-service tooling, and embedded security ownership across product engineering teams
  • NERC CIP v3 to v5 migration training and delivery as cybersecurity SME for SCADA/IoT environments (EY)
  • FISMA/FedRAMP ATO achieved within first year of building program from scratch (TPMC/SAIC)

Open to conversations about senior product security, AI/LLM security, and security program leadership roles.

LinkedIn Email

Popular repositories Loading

  1. Dome9toDD Dome9toDD Public

    Dome 9 to DefectDojo Generic Import Script

    Python

  2. Veritas-POC Veritas-POC Public

    Use an Agentic AI approach to code review using multiple stages. Build your gating criteria, start up a LM Studio instance or buy API tokens, and start doing first pass reviews!

    Python

  3. security-champions-kit security-champions-kit Public

    A practitioner-built Security Champions program framework covering the full 36-month lifecycle, from relaunch to institutionalization. Includes belt progression, playbooks, metrics, recognition, on…

  4. joshconkel joshconkel Public

    More About Me!

  5. mcp-sentinel mcp-sentinel Public

    A CLI tool that analyzes MCP server definitions (static) and optionally probes live MCP servers (dynamic) for security vulnerabilities specific to the agentic AI attack surface.

    Python