#### Neo: AI Security Engineer Overview

Neo is a cloud-based AI security engineer that works alongside your team, undertaking real security tasks similarly to a co-engineer. As it operates, it continuously learns your systems and processes, enhancing its capabilities over time.

**Key Features of Neo**  
- **Framework Design**: Neo combines large language model reasoning with execution tools, isolated sandboxes, and deep integrations into your organization. Users have visibility and control over its operations.
- **Performance Evaluation**: In a recent evaluation, Neo successfully completed a full case study for an L5 principal security engineer interview, demonstrating its efficiency in understanding environments, investigating issues, and proposing remediation plans.

**Current Use Cases for Enterprise Security Teams**  
1. **Vulnerability Backlog Triage and Closure**  
   - **Today**: Engineers spend hours manually managing incoming tickets.  
   - **With Neo**: Automated findings clustering, prioritization, and ticket updating streamline the process, allowing engineers to focus on decision-making.

2. **Pre-release Feature Reviews and Targeted Testing**  
   - **Today**: Security reviews can be shallow due to workload.  
   - **With Neo**: Automatic threat modeling, test plan creation, and issue documentation improve review thoroughness.
   
3. **Continuous Compliance and Audit Preparation**  
   - **Today**: Security teams scramble to assemble evidence for audits.  
   - **With Neo**: Periodic checks and evidence collection streamline the audit preparation process.

### How Neo Works
1. Define an objective (e.g., compliance review).
2. Neo gathers context and designs a multi-step plan.
3. Custom agents execute tasks while maintaining logs in real-time.
4. Artifacts like results and reports are saved for reuse.

**Key Components**  
- **Custom Agents**: Specialized agents tuned to specific security contexts.
- **Memory**: Maintains understanding of the environment, enhancing task execution.
- **Sandboxes and Tools**: Isolated environments for secure task execution minimizing risks to production systems.

### Safety, Transparency, and Control
- **Isolated Execution**: Tasks run in controlled environments to ensure safety.
- **Data Management**: User data remains private and is not used for model training.
- **Logging**: Every action is documented for user review and accountability.

### Availability and Target Users
Neo is currently in limited release, focusing on security engineering teams in large organizations that have previously experimented with tools like LLMs but faced limitations. It is designed for teams requiring safe AI systems to delegate security tasks effectively.