Field Guides / Breaking into technology / Language & NLP
Track D · last checked November 19, 2026
Language & NLP
You know a second language well, or you are fascinated by how meaning survives translation, and you want to work on how software handles language.
The work: Build and check language datasets, test how AI systems handle different languages, localize products, and, for engineers, build natural language processing (NLP) features.
Good fit if you’ve been: Military linguists and language analysts, translators, interpreters, language teachers, and anyone with tested proficiency in another language.
Tested language proficiency is a real differentiator in AI evaluation and localization, which need people who can judge output quality in a language. Keep your official test results (for example, DLPT scores) and describe your work in releasable terms only.
Try it first
Experiment (2–5 hours): Take 30 public sentences, compare two machine translations, and categorize the errors. If this is your first one, do the foundations first.
Learn in this order
| Learn | What it unlocks |
|---|---|
| 1. Python, Unicode, regular expressions, tokenization and text corpora | Build reliable multilingual datasets |
| 2. Classification, embeddings and transformers | Understand how models represent language |
| 3. Compare machine translation and retrieval by language and error type | Apply linguistic judgement to technical evaluation |
| 4. Publish an evaluated tool or a public multilingual test set | Evidence of a distinctive specialty |
Portfolio project: A public test set of about 100 examples in the languages you know, two models compared, results reported by language and error type, and notes on limitations, licensing and cost. You should be able to explain every part of it, reproduce the results and discuss its limits.
Credentials and degrees
Credentials: Proficiency evidence beats generic AI certificates here. For engineering roles, show Python and NLP projects. Security+ only for DoD roles that require it.
Degree: A language or linguistics degree plus programming for evaluation and localization; Computer Science for NLP engineering; computational linguistics graduate study for research. How to compare schools.
Where it leads
First jobs to search for: Localization QA, language data specialist, AI evaluator, language analyst; junior developer if you code.
Later: NLP engineer, computational linguist, multilingual AI specialist, localization engineer.
Reality check: Some AI evaluation and annotation work is short-term contract work. Check pay, hours and terms before you commit, and treat it as experience on the way to a staff role.
Related DoD work roles (orientation only; about DoD 8140): 151 Multi-Disciplined Language Analyst (billet-dependent); 623 AI/ML Specialist; 672 AI Test & Evaluation Specialist.
| Occupation | U.S. 25th percentile | U.S. median | Utah 25th percentile | Utah median |
|---|---|---|---|---|
| Interpreters and translators | $47k | $60k | $50k | $66k |
| Software QA analysts and testers | $80k | $104k | $62k | $82k |
| Software developers | $105k | $136k | $101k | $129k |
BLS May 2025 occupational wages via O*NET, all experience levels, rounded to $1,000. The 25th percentile is not a starting salary; first offers are often lower. No clearance, bonus or equity premium included. How to read these numbers.
Free places to start
Next moves
Enjoyed it? Stay and build the portfolio project. Enjoyed part of it more? Try a neighbor:
- AI assurance & governance: Evaluating accuracy, reliability and risk.
- AI / machine learning: Experimenting with models, code and mathematics.