LLM Engineer: AI Features for Safer Aviation

Beacon AI; San Carlos (CA); Hybrid; USD 120,000 - 180,000; Full time

Benefits offered by this job

Healthcare: 100% employee premiums

3 weeks PTO + 13+ holidays

401(k) with future enhancement

Job summary

Beacon AI in San Carlos, CA is seeking an engineer to ship LLM-powered features end-to-end. You will design retrieval and tool-calling flows, implement APIs in Python/TypeScript, and optimize latency and cost in production.

You’ll work with ML/infra and product teams, iterating on evals, guardrails, and safety policies in a fast-paced, safety-critical aviation context. This role is ideal for developers early in their ML career (2–4 years) who want to own features from idea to production and

Qualifications

• This role is for engineers early in their LLM/production ML career (2-4 years software engineering experience).
• You'll implement features within existing service patterns, with architecture and eval design guided by senior engineers.

Responsibilities

• Build user-facing LLM features with retrieval-augmented generation and tool-calling flows.
• Own the service layer with APIs and workers in Python or TypeScript.
• Retrieval and data prep collaboration for embeddings, chunking, and indexing.
• Evaluation and quality: offline evals, online metrics, A/B testing with guardrails.
• Safety, privacy, and compliance: content checks, PII redaction, access controls, auditing.
• Operate what you build: tracing, logs, dashboards, and debugging across components.

Skills

LLM features

Production code

Tests

Docs

Embeddings

Vector search

Tool-calling

Tools

Python

TypeScript

LangChain

OpenAI API

AWS Bedrock

OpenSearch

pgvector

Pinecone

Weaviate

S3

Aurora

DynamoDB

Job description

Beacon AI in San Carlos, CA is seeking an engineer to ship LLM-powered features end-to-end. You will design retrieval and tool-calling flows, implement APIs in Python/TypeScript, and optimize latency and cost in production.

You’ll work with ML/infra and product teams, iterating on evals, guardrails, and safety policies in a fast-paced, safety-critical aviation context. This role is ideal for developers early in their ML career (2–4 years) who want to own features from idea to production and