Principal Engineer – Tech Lead, Embodied AI, Off-Board Performance Evaluation

This is a fully remote position, open to applicants in United States.

📋 Description

• Provide technical oversight for the architecture aimed at identifying, detailing, and enhancing events in historical vehicle logs utilizing Multimodal LLMs.

• Supervise the off-board ingestion and integration of semantic scene descriptions, ego-centric kinematics, and internal autonomy telemetry to establish a comprehensive diagnostic context for LLM inference.

• Create structured prompting templates employing Contextual Prompting (CP), Chain-of-Thought (CoT), and In-Context Learning (ICL) to assess various scenarios.

• Design the integration of foundation models into the Metrics Engine (ME), formulating efficient cascade filtering and log slice parallelization strategies to scale high-volume LLM inference across simulation and on-road drive logs while controlling computational latency and costs.

• Define, design, and implement essential metrics to evaluate the performance of autonomous vehicles, such as lane change capabilities, oscillations, and braking effectiveness.

• Launch and manage a Retrieval-Augmented Generation (RAG) vector database that includes codified AV Driving Policies, grounding off-board LLM evaluations within specific Operational Design Domains.

• Act as a technical escalation resource and collaborate with Autonomy (Planner, Prediction, Perception) and Systems teams to provide high-signal, enriched event data.

• Lead the shift towards Direct Vector-LLM Fusion by utilizing emerging Physical AI ecosystems and open-weights Vision-Language-Action (VLA) models to process telemetry off-board, eliminating text-translation bottlenecks.

⛳️ Requirements

• Over 10 years of professional experience in software engineering, applied AI/ML, or the development of autonomous vehicle systems.

• A Bachelor's degree in Computer Science, Engineering, Robotics, or a related discipline.

• Demonstrated experience with Large Language Models (LLMs) and Vision-Language Models (VLMs) for reasoning, parsing, and scene description tasks.

• Background in parameter-efficient fine-tuning and deploying open-weights models on internal systems.

• Familiarity with local and cloud vector databases, such as LanceDB, for storing output vector embeddings.

• Experience in adversarial scenario generation and closed-loop simulation environments.

• Strong experience in using software to develop frameworks, libraries, and tools for calculating and aggregating AV performance metrics.

• Excellent analytical and problem-solving abilities, especially regarding complex system performance evaluation.

• Expert-level proficiency in Python along with a solid understanding of software development principles.

🏝️ Benefits

• Medical

• Dental

• Vision

• 401k with a company match

• Health saving accounts

• Life insurance

• Pet insurance

• More