AI Engineer
← All organizations

Application and AI Security

Snyk

Snyk helps developers and security teams find and fix vulnerabilities within development workflows. Snyk Code analyzes application source code, while Open Source checks dependencies for vulnerabilities and licensing risks. Container and Infrastructure as Code cover images and cloud configurations; API & Web tests running applications. Snyk Studio embeds security controls in AI coding assistants. Evo Continuous Offensive Security adds continuous AI-powered penetration testing and agent red teaming, while Snyk Secrets detects exposed credentials and provides prevention gates across coding and delivery tools.

Founded in 2015 by Guy Podjarny, Assaf Hefetz and Danny Grander, Snyk is now led by Ken MacAskill as interim CEO and CFO. Its developer-oriented approach includes Snyk Code’s combination of context-aware program analysis, a logic solver and human-guided machine learning, rooted in DeepCode technology acquired in 2020. The completed acquisition of Invariant Labs in 2025 extended its agent security research into tool poisoning and Model Context Protocol vulnerabilities, alongside guardrails for agent interactions.

Snyk reported serving over 4,800 customers globally in August 2026. In September 2025, it reported exceeding $300 million in annual recurring revenue, with Snyk Code contributing $100 million. Its financing history includes a $196.5 million Series G in 2022 at a $7.4 billion valuation.

snyk.io ↗

2 talks

Newest first

3 speakers at AIE

Affiliations reflect their AIE appearances, not necessarily current employment.

Ezra TanzerDirector, Product ManagementDan ArpinoSoftware engineerManoj NairCTO & Chief Innovation Officer

Start here

  1. Agentic Development Security

    Start with Tanzer and Arpino's talk to understand why scanning generated code needs to be accompanied by inspection of agent tooling and behavior.

    Ezra Tanzer · Dan ArpinoAI Engineer World's Fair 2026

  2. Through the AI Fog: The architectural decision the next 24 months of agentic security depends on.

    Use Nair's talk to consider how autonomous development changes remediation demands and the tools and training security engineers need.

    Manoj Nair · Ezra · RandallAI Engineer World's Fair 2026

Messages from the stage

Guardrails developers can work with

Arpino demonstrates an in-development local Electron assistant for monitoring agent sessions and addressing vulnerable tools. The session also examines the tension between strict security controls and developers' sensitivity to false positives.

Threats across skills, tools, and models

Nair's architecture discussion connects poisoned agent skills and MCP-connected tools with chained exploits and model-dependent data leakage. He closes by advocating stronger tooling and training for AI security engineers.

Affiliations reflect each recorded session, not necessarily current employment.

Company sources · checked 2026-08-27