AI Engineer: NLP & LLMs

Zyphe; San Francisco (CA); Hybrid; USD 120,000 - 150,000; Full time

Job summary

Zyphe is seeking an AI Engineer specializing in NLP and large language models to develop intelligent language systems for compliance automation. In this role, you will build and fine-tune large language models and design production pipelines.

The ideal candidate has extensive experience with transformer architectures and proven abilities in designing RAG systems. This is a remote position with a flexible hybrid policy.

Qualifications

• Strong experience building production NLP systems, not just prototypes.
• Deep understanding of transformer architectures and training dynamics.
• Hands-on experience with LLM fine-tuning and RLHF or DPO.

Responsibilities

• Build and fine-tune large language models for compliance analysis.
• Design and optimize RAG pipelines for regulatory knowledge bases.
• Implement inference optimization and develop prompt engineering frameworks.

Skills

Production NLP systems

Transformer architectures

LLM fine-tuning

Python

HuggingFace ecosystem

Education

Bachelor's degree in a related field

Tools

vLLM

TGI

Job description

About Zyphe

Zyphe is building agentic compliance for regulated fintechs, crypto exchanges, and enterprises. Our five-agent system (Document, Liveness, Name Resolution, Risk Profile, Historical) automates KYB, KYC, and AML while a privacy‑first architecture means personal data never has to live on our customers' servers.

We are early, focused, and shipping. Real customers, real revenue, an active fundraise with a committed lead. The next twelve months are about turning that into a defended category position.

About the role

Zyphe is hiring an AI Engineer specializing in NLP and large language models to build intelligent language systems that power our compliance automation and document understanding.

This is not a prompt‑engineering seat. You will own the full NLP stack, from fine‑tuning foundation models to designing production RAG pipelines and evaluation frameworks, working at the intersection of LLM engineering, information extraction, and regulatory intelligence.

Responsibilities

• Build and fine‑tune large language models for document classification, entity extraction, and compliance analysis.
• Design and optimize Retrieval‑Augmented Generation (RAG) pipelines for regulatory knowledge bases.
• Develop prompt engineering frameworks and evaluation harnesses for LLM‑powered features.
• Implement inference optimization (quantization, distillation, speculative decoding).
• Build structured output extraction from unstructured identity documents and compliance filings.
• Create automated evaluation pipelines for accuracy, hallucination rates, and latency.
• Translate regulatory requirements into AI capabilities with product and compliance teams.
• Stay current on the LLM landscape and evaluate new models and techniques.

You may be a good fit if you

• Have strong experience building production NLP systems, not just prototypes.
• Have a deep understanding of transformer architectures, attention mechanisms, and training dynamics.
• Have hands‑on experience with LLM fine‑tuning (LoRA, QLoRA, full fine‑tuning) and RLHF or DPO.
• Have a proven ability to design and optimize RAG systems at scale.
• Have experience with inference optimization and model serving (vLLM, TGI or similar).
• Have strong Python and HuggingFace ecosystem fluency.
• Understand evaluation methodologies for generative AI.

Strong candidates may also have experience with

• Information extraction from semi‑structured documents.
• Regulatory or legal NLP applications.
• Building eval harnesses and adversarial test suites for LLM features.
• Privacy‑preserving NLP and on‑device inference.
• Open‑source contributions to LLM tooling or eval frameworks.

Annual salary

Competitive, commensurate with experience. Equity included.

Logistics

Location: Remote. Hybrid policy: Fully remote, with periodic on‑sites for off‑sites and key meetings. Visa sponsorship: Not available at this time.

Education

We require at least a bachelor's degree in a related field, or equivalent professional experience.