עדיין מחפשים עבודה במנועי חיפוש? הגיע הזמן להשתדרג!
במקום לעבור לבד על אלפי מודעות, Jobify מנתחת את קורות החיים שלך ומציגה לך רק משרות שבאמת מתאימות לך.
מעל 85,000 משרות • אלפי משרות חדשות בכל יום
חינם. בלי פרסומות. בלי אותיות קטנות.
DOCX, DOC, PDF - בעברית או באנגלית
כבר העליתם קו"ח בעבר? כניסה
About Bay AI agents now write code, run commands and touch company data on employee laptops, in the browser, and in cloud sandboxes they spin up on demand. Bay is building the security layer that lets enterprises adopt them without losing control. Bay is an agentic security platform. It discovers every AI agent, MCP server, IDE and browser extension across a company's endpoints and the cloud environments they connect to. It separates human activity from agent activity and enforces real-time guardrails on what agents can do, with no persistent agent to install. Founded in 2025 by Yuval Zacharia (CEO), Aner Morag (CPO) and Boris Danilovich (VP Engineering), Bay has R&D in Tel Aviv and a growing go-to-market team in the US and Europe. Our team comes from Noname Security (acquired by Akamai), SentinelOne, Unit 8200, Mamram and other leading cyber companies, and we already protect enterprise customers in the US. The role We're hiring a Security Researcher to take apart the AI agents enterprises are adopting, find where they break, and work out how to control them. Today's agents rarely live in one place. A desktop or browser client sits on the endpoint, the model runs in the vendor's cloud, and the work often happens in a remote VM or container. For each agent, you'll map that chain: where actions execute, how commands travel, who is authenticated, and where we can observe or block. Sometimes the answer looks like offense, such as injecting into a process or speaking an agent's internal protocol, because that's where an enforcement point belongs. You'll be part of a research group we're building from scratch, working with our Head of Research, detection engineers and R&D. We want every project to do two things: make Bay's product better, and produce something worth publishing. What you'll do Research AI agents end to end: coding agents (Claude Code, Codex, Cursor, Copilot), cloud and computer-use agents, browser agents and extensions, and MCP servers, across the local client, the cloud control plane and the remote runtime. Reverse engineer clients, runtimes and protocols (Electron/Node apps, native and Rust binaries, sandboxed VMs and containers) on macOS, Windows and Linux to find where agent actions can be observed, intercepted or blocked. Hunt for vulnerabilities in high-impact agent products, such as sandbox and container escapes, cross-tenant access, credential exposure, prompt-injection-to-execution chains, and tool or MCP abuse. Take them through responsible disclosure. Turn findings into product: interception points, enforcement primitives and threat insight that R&D and our detection engineers build into Bay. Build research tooling and PoCs, using AI agents as research partners (LLM-driven Frida, IDA or Ghidra over MCP) to move faster. Track new agent releases, agentic threats, and malicious packages, skills and extensions in the wild. Publish blogs, CVEs, talks and open-source tools that build Bay's voice in the community. Red-team Bay's own guardrails as the team grows. What we're looking for A track record of original security research: vulnerabilities you found, disclosed or exploited (CVEs, bug bounty, vendor or red-team research). Usually 3+ years, but we care more about what you found than how long it took. Hands-on black-box research: dynamic instrumentation (Frida or similar), traffic interception and TLS decryption, and protocol analysis when you only have the client or only the server. Reverse engineering with IDA, Ghidra, Binary Ninja or similar, including stripped binaries, certificate pinning and custom transports. Strong OS internals: processes, IPC, sandboxing (namespaces, seccomp, micro-VMs) and networking. macOS is a plus. Strong coding in Python plus at least one of TypeScript/JavaScript, Go, Rust or C/C++. You build your own tools. Fluency with AI agents as research tools, and the judgment to check what they tell you. A nose for impact: you pick targets that matter, know when to push harder and when to move on, and work independently from question to result. Evidence discipline: you keep what you observed and reproduced separate from what you inferred. Nice to have Research on LLM or AI security, prompt injection, agent sandboxes or software supply chain. Cloud and container security: Kubernetes, container runtimes, cloud identity. Published research, conference talks or open-source tools. Background in an IDF technology or intelligence unit (8200, 81, Mamram, Sapir or similar). Exposure to detection engineering or threat hunting. B.Sc. in Computer Science, Mathematics or a related field.
במקום לעבור לבד על אלפי מודעות, Jobify מנתחת את קורות החיים שלך ומציגה לך רק משרות שבאמת מתאימות לך.
מעל 85,000 משרות • אלפי משרות חדשות בכל יום
חינם. בלי פרסומות. בלי אותיות קטנות.
DOCX, DOC, PDF - בעברית או באנגלית
כבר העליתם קו"ח בעבר? כניסה
שאלות ותשובות עבור משרת Security Researcher
כחוקר אבטחה ב-Bay AI, תהיה אחראי על ניתוח מעמיק של סוכני AI ארגוניים מקצה לקצה. המטרה שלך היא להבין כיצד סוכנים אלו פועלים, לזהות נקודות תורפה בתשתית שלהם ולפתח שיטות שליטה ואכיפה כדי להבטיח שארגונים יוכלו להשתמש בטכנולוגיות AI ללא סיכון, תוך כדי תרומה למחקר מקצועי ופרסום ממצאים לקהילה.
משרות נוספות מומלצות עבורך
-
Senior/Principal Security Researcher - Virtualization (Cortex)
-
תל אביב - יפו
Palo Alto Networks
-
-
network analyst Cyber security protection
-
תל אביב - יפו
צ'ק פוינט
-
-
Security Researcher
-
תל אביב - יפו
CYCL (Powered by Bugsec)
-
-
Data Protection Security Researcher
-
תל אביב - יפו
Jazz
-
-
🔍 Embedded Security Researcher | Reverse Engineering
-
תל אביב - יפו
Pacaya HR Experts
-
-
Security Researcher
-
תל אביב - יפו
Lema AI
-