Senior AI Developer

Boston Research Journals; Somerville, Northern (MA, KY); Hybrid; USD 120,000 - 190,000; Full time

Job summary

Boston Research Journals is building AI-assisted tooling to streamline scholarly publishing, including manuscript screening, metadata extraction, and editorial copilots. You will develop production features on LLM/ML APIs, ensuring measurable improvements through robust evaluation harnesses and clear build-vs-buy decisions.

The role emphasizes responsible AI use in editorial pipelines, with a focus on transparency, reliability, and publishing ethics.

Qualifications

• 4+ years of software engineering experience.
• 1–2 years building production features on LLM or ML APIs.
• Proficiency in TypeScript or Python.
• Hands-on experience with prompt engineering and Retrieval-Augmented Generation (RAG).
• Strong judgment on AI usage in editorial workflows.

Responsibilities

• Design and ship LLM-backed features for manuscript triage, metadata extraction, and editorial assistants.
• Build evaluation harnesses to measure model/prompts changes before editors see them.
• Integrate third-party detection/enrichment APIs; assess build-vs-buy decisions.
• Own cost, latency, and reliability of AI features in production.
• Ensure AI usage remains transparent and defensible in publishing ethics.

Skills

Software engineering

LLM/ML APIs

TypeScript

Python

Prompt engineering

RAG

Ediorial-pipeline judgment

Tools

Node.js

PostgreSQL

Job description

Build AI-assisted tooling for scholarly publishing - manuscript screening, similarity and AI-content detection, metadata extraction and editorial copilots.

About the Role

We are bringing modern AI into the day-to-day work of running peer-reviewed journals: screening incoming manuscripts, extracting structured metadata, assisting editors with summaries and correspondence, and keeping research integrity checks (plagiarism and AI-content detection) accurate and fair.

What You Will Do

• Design and ship LLM-backed features: manuscript triage, metadata extraction, editorial assistants and reviewer support tools.
• Build evaluation harnesses so every model or prompt change is measured before it reaches editors.
• Integrate third-party detection and enrichment APIs, and decide build-vs-buy with clear analysis.
• Own the cost, latency and reliability of AI features in production.
• Keep our use of AI transparent and defensible - publishing ethics matter deeply here.

What We Are Looking For

• 4+ years of software engineering, including 1-2 years building production features on LLM or ML APIs.
• Strong TypeScript or Python; comfort integrating with a Node.js/PostgreSQL backend.
• Practical prompt-engineering and RAG experience, and the habit of evaluating rather than eyeballing.
• Sound judgment about where AI should and should not be used in an editorial pipeline.

Nice to Have

• Experience with document parsing (PDF/LaTeX/DOCX) at scale.
• Familiarity with research-integrity tooling (iThenticate, Turnitin, AI-content detectors).
• Publications or side projects in NLP / information extraction.