Logo
ZipRecruiter

Lead Machine Learning Scientist - Agentic AI & Edge Security

ZipRecruiter, San Francisco, California, United States, 94199

Save Job

Job DescriptionJob Description

Position Details

Employment:

Full-Time

Citizenship Requirement:

U.S. or Green Card Holder

Security Clearance:

N/A

Travel:

To HQ as needed, on-site presence at client sites when required

Compensation:

Competitive Salary + Benefits Package

About RightSeat.AI

At RightSeat.AI, we are dedicated to transforming the future with advanced artificial intelligence. We assist companies in streamlining operations, enhancing customer experiences, and driving innovation. At RightSeat.AI, we promote a dynamic, inclusive work environment where every voice is valued. Whether you're starting your career or seeking new challenges, you'll have opportunities to grow, learn, and make a meaningful impact.

About Our Client

Our client is a Sequoia-backed cybersecurity startup developing a groundbreaking enterprise security platform purpose-built for the AI era.

Founded by entrepreneurs with multiple successful outcomes, the team brings deep expertise from leading security companies and has played key roles in developing Microsoft's Security Copilot and Microsoft Defender. Their previous ventures have led to major industry acquisitions and the creation of widely adopted threat intelligence solutions.

Now, our client is building its core R&D team to shape the future of AI- cybersecurity. This is a rare opportunity to help build a foundational company for the AI era, starting from day one.

Position Overview

We're hiring a

Lead Machine Learning Scientist

to architect and deploy intelligent, multi-agent systems for real-time threat detection and response at the edge. This role blends

applied AI research

with

production-grade engineering , ideal for someone who thrives in building scalable ML systems and wants to define the roadmap for AI-driven cybersecurity.

You'll lead the development of

agentic frameworks

using

OpenAI GPT-4

and

Gemma 3N , fine-tune LLMs for security-specific tasks, and build

low-latency inference systems

for edge deployment. Your work will directly shape how enterprises detect, investigate, and respond to threats in dynamic environments.

We're especially interested in candidates who have worked on or are inspired by projects like:

Claude CUA

– Autonomous agents with vision and reasoning (trycua.com)

OpenAI Operator

– LLM-powered automation and orchestration

Gemini Vision

– Multimodal perception and interaction

Microsoft Phi-Vision

– Lightweight vision- models

Microsoft Agent Arena

– Multi-agent simulation and benchmarking

Omniparser

– Structured data extraction from unstructured formats

E2B Desktop Sandbox

– Secure graphical environments for LLMs (GitHub)

Key Responsibilities

Architect agentic AI systems using frameworks like AutoGen, LangGraph, and Semantic Kernel.

Build multimodal models that combine vision and for autonomous reasoning and secure execution.

Develop sandboxed desktop environments for LLMs using E2B and similar platforms.

Fine-tune LLMs for structured data extraction, threat detection, and endpoint telemetry analysis.

Engineer scalable pipelines for ingesting and parsing system logs, telemetry, and visual data.

Optimize models for low-latency inference and edge deployment using ONNX, TensorRT, and quantization.

Collaborate with engineering teams to integrate AI into endpoint security workflows with a focus on data integrity, interpretability, and robustness.

Define metrics and adversarial testing protocols for evaluating model performance in security-critical environments.

Required Qualifications

10+ years of experience in machine learning, AI systems, or computer vision.

Deep expertise in generative models, multimodal architectures, and data-centric AI.

Hands-on experience with agentic orchestration frameworks and platforms like Claude CUA, Gemini Vision, and E2B Desktop.

Hands-on experience with OpenAI GPT-4, Gemma 3N, and agentic orchestration frameworks.

Strong background in endpoint security, telemetry analysis, and edge computing.

Proficiency in Python, PyTorch, Hugging Face Transformers, and LangChain.

Experience deploying models to edge environments with performance and security constraints.

Familiarity with secure MLOps, CI/CD pipelines, and infrastructure-as-code (e.g., Terraform, Helm).

Attributes

Entrepreneurial mindset with a passion for building from the ground up.

Strong communication skills and ability to collaborate across disciplines.

Experience mentoring or leading technical teams.

Interest in shaping company culture and technical direction.

Why Work for Our Client

Founding Role : Help define the future of agentic AI and data-driven security.

Mission-Driven : Build AI that protects people and organizations.

Elite Team : Collaborate with pioneers from Microsoft, Uber, and ServiceNow.

Fast Iteration : Ship impactful features quickly.

Equity & Impact : Share in the upside of a Sequoia-backed startup.

Our Clients Benefits

Team members enjoy a comprehensive benefits package that supports their well-being. You'll share in the company's long-term success as part of a venture-backed startup. Benefits include health, dental, and vision coverage, plus flexible paid time off to promote a healthy work-life balance.

Equal Opportunity Employer Statement

RightSeat.AI is dedicated to fostering an inclusive and equitable workplace. We make employment decisions based on qualifications, merit, and business needs, regardless of , , , , veteran status, or any other protected characteristics.