Source Register and Validation

Research pass: 2026-10-06. Scope: the new CS curriculum, foundational math/data books, and the identity/access of the AI engineering book. This is not a completed audit of every claim, link, benchmark, or implementation in the repository.

Evidence and access

Prefer the author’s, publisher’s, university’s, or standards body’s own page. A free companion repository is not a free textbook. “Free to read” does not imply permission to redistribute or relicense. Consult the source’s license before copying content; this register links to resources rather than reproducing them.

“Checked” below means the cited official page or document was readable during this pass and supported the listed identity or scope. It does not mean every exercise, download, chapter, or factual statement was independently verified. Live course pages can change; record the semester/version when using them.

Mathematics and data books

Source	Author / edition	Access and evidence	Curriculum use
Calculus, Volumes 1-3	Gilbert Strang and Edwin Herman, senior contributing authors	Checked official preface: free web/PDF; lists scope of all three volumes	Single-variable differentiation/integration, series, multivariable/vector calculus
Linear Algebra	Jim Hefferon	Checked author page: free text and supporting resources	Systems, vector spaces, maps, eigenvalues; pair with numerical labs
Book of Proof	Richard Hammack, third edition	Checked author page; older VCU URL failed retrieval	Logic and proof before algorithm correctness
Discrete Mathematics: An Open Introduction	Oscar Levin, fourth edition	Checked official book page; free online	Counting, sequences, logic and graphs; match exercises to edition
Introduction to Probability	Charles M. Grinstead and J. Laurie Snell	Checked Dartmouth-hosted full PDF	Discrete/continuous probability and stochastic reasoning
Introductory Statistics 2e	Barbara Illowsky and Susan Dean	Checked publisher preface; free web/PDF	Applied inference and regression; algebra-based, not a mathematical-statistics replacement
Mathematics for Machine Learning	Marc Peter Deisenroth, A. Aldo Faisal, Cheng Soon Ong; 2020	Checked authors’ companion site with free PDF	Bridge from linear algebra, calculus, probability and optimization to ML
An Introduction to Statistical Learning	James, Witten, Hastie, Tibshirani; Taylor also on Python edition	Checked authors’ site: R second edition (2021), Python edition (2023), free PDFs	Regression, classification, resampling and statistical learning labs
Deep Learning	Ian Goodfellow, Yoshua Bengio, Aaron Courville	Checked official book site; free HTML	Mathematical foundations and neural networks
Designing Data-Intensive Applications	Use author/publisher metadata for the chosen edition	Checked official landing page; commercial book	Storage, replication, transactions and distributed data; existing chapter references must be checked against the edition used
Fundamentals of Data Engineering	Joe Reis and Matt Housley	Checked publisher listing; paid/subscription full book	Data lifecycle, architecture and operational decisions; optional companion to local docs/course routes

Computer science foundations

Source	Owner / author	Access and evidence	Curriculum use
CS2023 and knowledge areas	ACM / IEEE-CS / AAAI	Checked official index and endorsement metadata	Breadth cross-check; not evidence of accreditation or full outcome alignment
Composing Programs	Official textbook site	Checked root, which redirects readers to third edition; detailed edition content not audited	Programming foundations
Open Data Structures	Pat Morin	Checked author site; free book	Data structures before advanced algorithm analysis
Algorithms	Jeff Erickson, first edition, 2019	Checked author page; free electronic book and course archives	Proof-based algorithms; explicitly assumes discrete math and basic data structures
Nand to Tetris	Noam Nisan and Shimon Schocken	Checked official course: free project/lecture route; book and certificates are separate	Architecture and hardware/software construction
Operating Systems: Three Easy Pieces	Remzi H. and Andrea C. Arpaci-Dusseau	Checked authors’ site; free chapters	Virtualization, concurrency, persistence
An Introduction to Computer Networks	Peter L. Dordal	Checked university-hosted book	Network fundamentals before application protocols
CS186	UC Berkeley	Checked course page; public materials, no assumed access to enrolled-student grading	Database foundations and internals
6.045J, Spring 2011	MIT OpenCourseWare	Checked archived course	Automata, computability and complexity
CS161 textbook	UC Berkeley	Checked public textbook	Security foundations
6.831, Spring 2011	MIT OpenCourseWare	Checked archived course	HCI and usability methods; use current platform accessibility guidance for implementation
CS184, Spring 2026	UC Berkeley	Checked semester-specific course page	Computer graphics

Chip Huyen’s books

The likely intended reference is Chip Huyen, not “Nyugen”: AI Engineering: Building Applications with Foundation Models, 2025. The official repository identifies the author/year and links to purchase options, chapter summaries, notes and the table of contents. These are free supporting materials, not the complete commercial book.

The same author describes Designing Machine Learning Systems as a companion focused on traditional ML applications, feature engineering and training. Use AIE for foundation-model applications; use DMLS for the broader production ML lifecycle. Both are optional paid references here.

Reading theme	Local application	Evidence to produce
Evaluation	LLM evaluation	Frozen test set, baseline, error taxonomy, uncertainty
Retrieval and context	Retrieval & RAG	Retrieval metrics and end-to-end answer assessment separately
Data quality	Data engineering	Provenance, deduplication and leakage checks
Inference and application architecture	LLM systems	Quality, latency, throughput and cost under a stated workload

Cross-validation protocol

For future source additions, capture: stable source URL, title, authors, edition/date, access category, topic, prerequisite, specific claim or learning outcome, check date, and reviewer. For technical claims, record a section/page or named result and assumptions. Independently recompute quantitative examples or compare with an independent implementation. Two model answers agreeing is not independent evidence when both repeat the same source.

For datasets used in exercises, also record origin, license, collection dates, version/hash, units, schema, missingness, sampling, transformations, train/test split rules and known limitations. Keep raw inputs separate from derived data. Check leakage, duplicate records, out-of-range values, and whether the sample supports the stated conclusion.

When another system reviews a claim, exchange a concrete record:

claim_id:
local_file_and_section:
claim_or_learning_outcome:
primary_source_and_locator:
edition_or_dataset_version:
independent_check_and_result:
disagreement_or_limitation:
reviewer_and_date:
status: pending | supported | corrected | unresolved

Findings and remaining work

Finding	Resolution / status
AI Engineering attribution and access	Official repository supports Chip Huyen (2025); paid text and free companion materials distinguished
Degree breadth versus interview tracks	Added explicit core sequence, external courses, coverage boundaries, and capstone requirements
Math workload	Paired-course schedule now budgets hours per subject; intensive track is distinguished from full semester study
Free access versus open license	Removed the math index’s blanket assertion that every linked resource is openly licensed
Other-system coordination	Pending: system identity/contact or review output has not been supplied; no second-system agreement is claimed
Full repository correctness	Pending: existing technical claims, time estimates, outdated links and edition-specific chapter references still need individual audits

Retrieval failures during research: the complete CS2023 HTML exceeded the browser’s content-size limit; the older Hammack VCU page returned 502. The official CS2023 index and Hammack’s current author page were used instead. Crafting Interpreters and a Cornell course URL could not be retrieved and were not promoted to checked sources. These retrieval failures do not establish that the websites are unavailable to other readers.