Editorial Standards

This page explains who writes the MemX blog and glossary, how that content is produced, and why we publish it. We want you to be able to judge our work on its merits, so we are direct about our process, including where we use AI tools and where humans make the decisions.

Who writes our content

Every article carries a real, named author with a public profile, not a generic “admin” or “team” byline. Our two primary authors are Aditya Kumar Jha, a software engineer at Neural Forge Technologies and a published author, and Arpit Tripathi, founder of Neural Forge Technologies. Both build MemX, the company's AI memory product. Technical and engineering topics are generally written under Aditya's byline; product, memory, and privacy topics under Arpit's. You can see all contributors on our authors page.

How we create our content

We use AI tools to assist with research and first drafts. Stating that plainly is deliberate: it does not lower the bar we hold the work to, because the editorial judgment, the sourcing, and the final accuracy are human responsibilities. Every article goes through the following process before it is published.

• Human direction. A human author chooses the topic, the angle, and the specific point the piece needs to make. We research real search demand and existing coverage so that each article adds something the current top results do not already say.
• AI-assisted drafting and research. We use AI tools to gather sources, structure arguments, and produce a first draft. AI is a tool in this process, not the author of record.
• Human fact-checking against primary sources. Every factual claim, date, statistic, version number, and citation is verified against a primary or authoritative source. We open each source link to confirm it is live and that it actually supports the claim. We do not publish a number, date, or quote we could not verify, and we do not guess a replacement when a source falls short.
• Editorial review. Drafts are edited for accuracy, clarity, and plain language. We cut filler, avoid hype, and write for a reader who wants the real answer, not a sales pitch.

Why we publish

MemX is an AI memory product, so we write about AI, large language models, memory, retrieval, and privacy. The goal of our blog and glossary is to explain these topics accurately and in plain language, for readers trying to understand them, not to manipulate search rankings. When a topic touches our product, we say so and keep the claim honest. We would rather lose a reader to a careful caveat than keep one with an overstatement.

Sourcing standards

We cite named, primary sources wherever possible: official documentation, the original research paper (with its arXiv or DOI identifier), the company that shipped a feature, or a recognized standards body. We prefer the source that created the fact over a blog that repeated it. Citations are linked so you can check them yourself.

Accuracy and corrections

We make mistakes, and when we find one we fix it. We periodically re-audit older articles because facts in this field change quickly: model versions, prices, and feature availability can all go stale. If you spot something that is wrong or out of date, email [email protected] and we will review and correct it. Material corrections are reflected in the article's last-updated date.

What we do not do

• We do not publish fabricated statistics, invented studies, or made-up personal anecdotes.
• We do not present unverified claims as fact, and we flag figures that are vendor-reported rather than independently audited.
• We describe MemX's privacy as private by architecture, with per-user isolation, encryption at rest, and on-device options. We do not claim end-to-end encryption or zero-knowledge, because that would not be accurate.
• We do not accept undisclosed paid placements in our editorial content.