AI Software Engineer
Alice (Formerly ActiveFence) · Ramat Gan, Tel Aviv District, Israel
- On site
- Full-time
- $200,000 / year
- Ramat Gan, Tel Aviv District, Israel
Job highlights
- Build AI safety environments for leading AI labs.
- Own end-to-end RL Gym platform development.
- Lead a team with high autonomy.
- Integrate systems with customer AI pipelines.
- Advance AI safety with novel solutions.
About the role
Principal Software Engineer, AI Safety Platforms
Alice's Innovation team is at the forefront of AI safety, developing adversarial reinforcement learning (RL) environments that train the world's most advanced AI models to be safer. Our clients, leading frontier AI labs, utilize these environments for post-training RL and safety evaluations. This role is pivotal in shaping the future of AI safety by creating systems that help next-generation models resist adversarial attacks.
Responsibilities:
As a Principal Software Engineer, you will own the RL Gym platform end-to-end. This includes architecting multi-site web environments simulating real-world attack surfaces and optimizing our in-house orchestration harness, AgenticVerse, for high-performance delivery into customer training pipelines. This is a builder role where you will lead a small team with high autonomy, moving rapidly from concept to production. You will also engage directly with customer engineering teams to understand their infrastructure constraints and ensure our environments meet their scale and reliability requirements.
Platform & Performance:
- Own and evolve AgenticVerse, our in-house orchestration harness for provisioning and managing RL environments at scale. Focus on performance, including low-latency provisioning, high concurrency, and minimal overhead per instance.
- Design and build isolated, reproducible web environments using technologies like Firecracker microVMs or Docker containers.
- Architect multi-site scenarios involving interconnected web applications with rich interactions such as drag-and-drop, file uploads, authentication, and LLM-in-the-loop components.
- Implement deterministic verifiers for unambiguous evaluation of agent behavior.
Customer Delivery:
- Collaborate directly with engineering teams at leading AI labs to integrate RL Gym environments into their training and evaluation pipelines.
- Translate customer specifications into working environments, iterating rapidly based on feedback.
- Manage the technical relationship, including SLAs and API contracts.
- Adapt environment delivery formats to customer infrastructure, supporting real-time API calls, offline batch processing, and various deployment models.
- Build customer-facing UIs, such as dashboards, configuration portals, and monitoring interfaces, as needed.
Rapid Prototyping:
- Transform ambiguous problem descriptions into working prototypes rapidly.
- Validate new environment types, interaction patterns, and verifier approaches quickly.
- Develop internal tooling to accelerate scenario authoring and testing.
Requirements:
Must-Have:
- 8+ years of software engineering experience with a proven track record of building production systems from inception.
- Deep expertise in infrastructure, including Linux, containers (Docker), VMs (Firecracker or similar), networking, and cloud platforms (AWS preferred).
- Strong Python skills and experience with async/concurrent systems.
- Experience building platforms or developer tools.
- Full-stack capabilities: backend services, infrastructure-as-code, APIs, and frontend development (React or similar).
- Demonstrated ability to work autonomously and make sound architectural decisions under ambiguity.
- Comfort working directly with external customers and translating technical constraints into solutions.
- English fluency (written and verbal).
Nice-to-Have:
- Experience with reinforcement learning infrastructure, training pipelines, or evaluation frameworks.
- Background in security, adversarial testing, or trust & safety systems.
- Familiarity with browser automation, headless browsers, or large-scale web scraping.
- Experience with Kubernetes operators or custom schedulers.
- Prior experience in a 0-to-1 environment (startup, innovation lab, or R&D team).
About Alice:
Alice is a leading trust, safety, and security company for the AI era, safeguarding communicative technologies. In a world transformed by AI, Alice provides comprehensive solutions across the AI lifecycle, supporting frontier model labs, enterprises, and UGC platforms with advanced evaluations, red-teaming, guardrails, and drift detection.
Key skills/competency:
- AI Safety
- Reinforcement Learning
- Platform Engineering
- Infrastructure
- Python
- Docker
- AWS
- System Architecture
- Customer Integration
- Software Development
Skills & topics
- AI Software Engineer
- AI Safety
- Reinforcement Learning
- Platform Engineering
- Python
- AWS
- Docker
- Infrastructure
- System Architecture
- Full Stack Development
- Principal Engineer
- MLOps
How to get hired
- Research Alice's Mission: Understand their focus on AI safety, trust, and security, and how this role contributes to their innovative solutions.
- Tailor Your Resume: Highlight your 8+ years of experience in production systems, deep infrastructure knowledge (Linux, Docker, AWS), Python, and full-stack development. Emphasize experience building platforms or developer tools.
- Showcase Autonomy & Customer Skills: Provide examples of working autonomously with minimal specification, making architectural decisions under ambiguity, and collaborating directly with external customers.
- Prepare for Technical Deep Dives: Be ready to discuss your experience with infrastructure, containerization (Docker, Firecracker), cloud platforms (AWS), Python concurrency, and building scalable platforms.
- Demonstrate Problem-Solving: Articulate your approach to rapid prototyping and translating complex, ambiguous problems into working solutions.
Technical preparation
Master Python for async and concurrent systems.,Deepen Linux, Docker, and AWS expertise.,Practice building scalable platforms and tools.,Prepare for system architecture design discussions.
Behavioral questions
Describe building a production system from zero.,How do you handle ambiguity in projects?,Share an experience collaborating with external customers.,How do you ensure system performance and reliability?
Frequently asked questions
- What is the primary focus of the Principal AI Software Engineer role at Alice?
- The primary focus is to own and evolve the RL Gym platform, building adversarial RL environments that train AI models for safety. This involves architecting web environments, optimizing the AgenticVerse orchestration harness, and ensuring high-performance delivery to customers.
- What kind of technical challenges can I expect as a Principal AI Software Engineer at Alice?
- You can expect deep technical challenges in platform engineering, infrastructure optimization (Linux, containers, cloud), high-performance systems, architecting complex web environments, and implementing deterministic verification methods. You'll also tackle customer integration and rapid prototyping.
- What makes this Principal AI Software Engineer role at Alice unique?
- This role is unique because your code directly influences how next-generation AI models are trained and made safer. You'll be at the cutting edge of AI safety, working in a field with no existing playbook and shaping its future.
- What level of autonomy can I expect in this Principal AI Software Engineer position?
- You will operate with high autonomy, leading a small team and driving projects from concept to production. The role is designed for builders who can make sound architectural decisions with minimal specification.
- How does Alice approach AI Safety?
- Alice is dedicated to AI safety, trust, and security. They provide end-to-end coverage across the AI lifecycle, from model hardening and red-teaming to runtime guardrails and drift detection, specifically for frontier AI labs and enterprises.
- What are the key technologies used by the AI Safety Platforms team at Alice?
- Key technologies include Python, asynchronous/concurrent systems, Linux, containers (Docker, Firecracker), cloud platforms (AWS), web frameworks (React), and potentially Kubernetes. Experience with RL infrastructure is also beneficial.
- What is AgenticVerse and what is its role in this Principal Software Engineer position?
- AgenticVerse is Alice's in-house orchestration harness. The Principal Software Engineer will own and evolve it, focusing on optimizing it for high-performance provisioning and management of RL environments at scale.
- How does Alice support customer integration for its RL Gym environments?
Similar roles
Open positions we recommend based on this role.