Principal AI Researcher

ProRata.ai•Bellevue, WA

10d•Onsite

About The Position

We’re looking for a Principal AI Researcher to join our Attribution Team, where you’ll design and develop algorithms and solutions for source attribution and hallucination detection. In this role, you will work closely with ML engineers, research scientists, and product teams to power answer analysis, source attribution, and compensation models for millions of users with fast, accurate, and comprehensive services. As Principal AI Researcher, you’ll be leading efforts to combine Natural Language and Multimodal Content Understanding, LLMs, search, reasoning and inference to build efficient, performant and scalable solutions for source attribution. This is a high-impact role focused on AI/ML, Natural Language Inference, and significant ownership in both technical decision-making and long-term architecture.

Requirements

PhD degree in AI or related field, or equivalent practical experience.
8+ years of experience in software engineering, with a focus on NLP/NLU, Symbolic and/or neural networks & reasoning topics, AI/ML and distributed systems.
Hands-on experience building and deploying content understanding and inference systems.
Deep knowledge of Entity, Event, Proposition extraction, NLI, KGs.
Experience with Search (OpenSearch, etc.) & Graph (Neo4J, etc.) databases.
Strong coding skills in Python and/or other backend languages (e.g., Rust, Java).
Experience with vector search, embedding pipelines, and dense retrieval techniques.
Excellent problem-solving, analytical, and debugging skills.
Strong sense of ownership, ability to work independently, and a self-starter mindset in fast-paced environments.
Passion for building impactful technology aligned with our mission.

Nice To Haves

Experience working on high reliability, observability, and monitoring of deployed systems is a plus
Proficiency in PyTorch, JAX, or other deep learning framework.
Experience with various LLM post-training techniques.
Significant experience with semantic search, embedding models, and prompt engineering.
Publications at top-tier AI conferences (e.g., NeurIPS, ICML, CVPR, ICLR).
Experience with proof, justification or compliance use cases.
Previous work on systems handling millions of users or queries per day.
Familiarity with cloud infrastructure (AWS, GCP, or Azure) and containerization tools (Docker, Kubernetes).

Responsibilities

Design, build and scale algorithms for content understanding, inference, entailment, relationship mining, retrieval, etc. in production-grade attribution stack.
Post-train foundation models (LLMs, VLMs) for content understanding and attribution solutions. Prototype, validate, and iterate rapidly on solutions to problems related to content understanding and source attribution.
Develop efficient extraction and indexing pipelines for deep understanding and representation of content for entailment and source verification in AI generated answers and other NLU scenarios.
Contribute to optimizing performance and latency for both real-time and batch use cases.
Support identifying and addressing bottlenecks in the attribution stack to improve response times and system efficiency.
Collaborate with cross-functional teams to integrate LLMs and attribution components into B2B and B2C services / applications.
Evaluate and integrate modern frameworks and tools to accelerate development.
Guide architectural decisions, mentor team members, and uphold engineering excellence.
Author technical papers or patents.