SnowPro Gen AI Specialty (GES-C02): Domain Weights and a 10-Week Study Path for a Snowflake-DE Candidate
The question
What are the SnowPro GenAI Specialty (C02) exam domain breakdown, weighting, and recommended 10-week study path for a candidate with strong Snowflake data engineering background? Context: target exam 2026-08-24 (~10 weeks out), the first cert escalator for the phData +$5k base raise; a flashcard deck exists but no structured C02 domain-weight breakdown or study-path sequencing is anchored.
What we already know (from the vault)
- A full study project already exists at [[readme]] (
01-projects/certifications/snowpro-genai-c02/) with per-domain notes, [[study-plan]], [[flashcards-deck]], [[third-party-resources]], and [[study-guide-summary]]. The structure is good; the domain-weight table in the README is mis-framed (see Contradictions). - [[study-plan]] currently recommends a 7-10 day sprint (Track A) premised on an early first-mover exam right after the May-2026 launch. That premise is now stale: deadline is 2026-08-24, so we have ~10 weeks, not 10 days. The sprint's content (RAG hands-on, function discrimination, RBAC/cost focus) is still correct; only the cadence needs re-keying.
- [[study-guide-summary]] confirms the official C02 study-guide PDF is auth-gated (HTTP 530 on direct fetch) — founder must pull it from
cp.certmetrics.com/snowflake. Official C02 domain weights were never published; C01 weights are our proxy. - [[cert-progress]] is the canonical tracker read by
/cert-study: Cert 1 (C02) has no study-log entries, no exam scheduled, no track selected as of 2026-06-13. Cert 2 (Anthropic) is far ahead (83% cold practice) and may get sat first. - [[third-party-resources]] already ranks the highest-value reads: Kratika Sharma's practical prep plan (#1), Eylon Steiner's official Snowflake Builders tutorial, plus DE-background pass accounts.
What the web says
- Four-domain C01 structure (the canonical pre-Feb-2026 weighting), confirmed across multiple sources including the official-channel Eylon Steiner tutorial on medium.com/snowflake: Domain 1 "Snowflake for Gen AI Overview — 26%"; Domain 2 "Snowflake Gen AI & LLM Functions — 40%"; Domain 3 "Snowflake Gen AI Governance — 22%"; Domain 4 "Snowflake Document AI — 12%."
- Feb-28-2026 update collapsed it to three domains (Document AI decommissioned), rebalancing to: Overview 30%, Gen AI & LLM Functions 44%, Governance 26% (Sharma, "Feb 2026 Update Included"). Sharma's key note: removing Document AI does not cut study time — reinvest into functions and governance "where the real detail lives."
- GES-C02 is the current exam code, successor to the retired GES-C01 (Snowflake learn page + voucher vendors confirm C02). C02 reinstated document processing but as functions (PARSE_DOCUMENT / AI_PARSE_DOCUMENT pipelines), not the legacy Document AI template-UI product. Official C02 percentage weights are not publicly published — treat any number as a proxy.
- Domain 2 is the make-or-break domain (40-44%). Sharma dedicated a full week to it: AI_COMPLETE, AI_CLASSIFY, EXTRACT_ANSWER, AI_PARSE_DOCUMENT, AI_SENTIMENT, AI_SUMMARIZE_AGG, AI_TRANSLATE, AI_EMBED, AI_COUNT_TOKENS, vector-similarity functions; plus Cortex Search, Cortex Analyst, Cortex Agents, fine-tuning, Copilot.
- The two most under-prepared, heavily-tested areas per every pass account: (1) RBAC / privileges — the exam "loves permissions"; know
CORTEX_USERdatabase role,CORTEX_MODELS_ALLOWLISTaccount param, "Required Privileges" doc sections, avoid ACCOUNTADMIN in labs; (2) cost model — not prices but what gets billed (token vs warehouse-credit vs Cortex Search multi-component billing), visible inACCOUNT_USAGEusage views. - Highest-ROI hands-on (universal recommendation): build one end-to-end project — Cortex Search service on unstructured text → Cortex Analyst on structured → combine into a Cortex Agent → drive with AISQL functions → watch tokens/credits land in usage views. "If you do one practical thing during prep, make it this."
- Exam mechanics (C01-inherited, C02 unconfirmed): ~55 questions, 85 min (~93 sec/q), scaled pass 750/1000, $375/attempt, online-proctored or test-center, multiple-choice + multiple-select + interactive scenarios.
Convergences and contradictions
- Convergence: Independent sources (official Snowflake Builders blog, Sharma, quick-facts pages) agree exactly on the C01 four-domain weights (26/40/22/12) and on the Feb-2026 three-domain rebalance (30/44/26). High confidence in relative emphasis: Functions ≈ 40-44%, Overview ≈ 26-30%, Governance ≈ 22-26% dominate; document processing is the lightest.
- Contradiction to fix in the vault: The README's domain table is wrong. It lists Domain 3 as "Building and fine-tuning models with SPCS + Model Registry (22%)" and Domain 4 as "document processing (12%)." Every external source — and the spirit of the C02 scope statement — puts Governance at the 22-26% slot, not SPCS. SPCS/Model Registry/fine-tuning are sub-topics that live inside the Overview and Functions domains, not a standalone weighted domain. Action: correct the README table (separate cleanup task; flagged here).
- Open contradiction (unresolvable from public web): C02's actual published weights and whether document processing is a 4th weighted domain or folded into Functions. Only the auth-gated official C02 study guide resolves this. Until founder downloads it, plan to the 3-domain post-Feb-2026 shape (30/44/26) and treat document-processing functions as part of the Functions block — that is the safest allocation and over-prepares the lightest area at low cost.
Synthesis for RDCO
Re-key the existing Track-A content to a 10-week cadence, front-load a diagnostic, and spend weeks proportional to weight — not to novelty. The candidate is a strong Snowflake DE (SQL, pipelines, warehousing, RBAC mechanics, ACCOUNT_USAGE views), so the Snowflake-platform substrate of every domain is already known. The genuinely new surface is the Cortex/LLM function catalog, RAG/agent assembly, and the Gen-AI-specific governance/cost nuances. Don't re-teach warehousing; skim it and pour hours into the 40-44% Functions domain and the permission/cost traps. Net effort target: ~3-5 hrs/week for 8 working weeks, with a diagnostic up front and a taper at the end — comfortably inside 10 weeks with slack for the day job.
Weeks 1-2 (Diagnose + Scope): Week 1 — founder logs into the learning portal, downloads the official C02 study guide PDF, registers for an exam slot (lock 2026-08-24 or earlier), and resolves the weights question against the official guide (then patch the README). Take one practice exam cold (pretesting effect; given the DE background and daily Cortex exposure at phData, the baseline will be high and will triage what to skim vs. drill — same diagnostic-first move that worked for Cert 2). Week 2 — capability map of the platform: Cortex Search, Analyst, Agents, Copilot, the AISQL/Cortex namespaces, and where SPCS/Model Registry/fine-tuning fit. This is mostly naming and boundaries, fast for an experienced DE.
Weeks 3-6 (The 40-44% core — Functions, RAG, Agents): This is the load-bearing block; give it four weeks. Weeks 3-4 — drill the function catalog until name-discrimination is automatic (AI_COMPLETE vs COMPLETE, AI_EXTRACT vs PARSE_DOCUMENT vs EXTRACT_ANSWER, AI_AGG vs SUMMARIZE_AGG, embedding dimensions 768 vs 1024 traps), running every example live in a personal sandbox (never client data). Week 5 — build the one canonical end-to-end project (stage PDFs → PARSE_DOCUMENT → Cortex Search service → Cortex Analyst → combine in a Cortex Agent → query with AI_COMPLETE), and watch tokens/credits appear in ACCOUNT_USAGE. Week 6 — document-processing functions + fine-tuning/Model Registry/SPCS at concept level (GPU = the SPCS-vs-warehouse tell). The DE muscle makes the plumbing trivial; the points are in knowing which function is best-practice for a scenario.
Weeks 7-8 (Governance, RBAC, Cost — the silent 22-26% that fails DEs): Two dedicated weeks because this is where pass accounts say candidates lose, and where DE confidence is a trap (the candidate knows generic Snowflake RBAC but not the Gen-AI-specific grants). Drill CORTEX_USER role, CORTEX_MODELS_ALLOWLIST, Cortex Guard, cross-region inference governance, and the billing-driver model (token vs credit vs Search multi-component). Practice all labs under a dedicated non-ACCOUNTADMIN role. Weeks 9-10 (Consolidate + sit): topic-based quizzes (not full mocks) on weak areas, second practice exam as a readiness gate, run the flashcard deck twice, rebuild the RAG pipeline from memory without copy-paste, then sit the exam. Light flashcard review morning-of; learn nothing new in the final 48 hours.
Open follow-ups
- Founder downloads the auth-gated official C02 study guide PDF and confirms: (a) actual C02 domain count and weights, (b) whether document processing is a standalone weighted domain or folded into Functions, (c) current question count / time / passing score (verify vs C01-inherited 55q/85min/750).
- Correct the mis-framed domain-weight table in [[readme]] (Domain 3 = Governance ~22-26%, not SPCS/Model Registry) — separate vault-hygiene task.
- Update [[study-plan]] Track-A cadence from "7-10 day sprint" to the 10-week shape above (or add a "Track C: 10-week deadline-keyed" section); current plan's premise (early first-mover launch window) is stale.
- Sequencing decision vs Cert 2: founder is at 83% cold on the Anthropic cert and could bank that +$5k first. Confirm which cert he sits first so the C02 study weeks are scheduled around it (Cert 2 escalator is the natural "pass within first 3 months" candidate).
- Confirm a clean personal Snowflake sandbox (trial or phData-provisioned non-client) for the hands-on labs to avoid touching client data.
Related
- [[readme]] — SnowPro Gen AI C02 study project root (domain table needs correcting)
- [[study-plan]] — existing Track A/B plan to re-key to 10 weeks
- [[study-guide-summary]] — official scope statement + auth-gated PDF location
- [[third-party-resources]] — ranked external prep reads (Sharma, Steiner, DE pass accounts)
- [[flashcards-deck]] — existing Anki-importable Q&A deck
- [[cert-progress]] — canonical /cert-study tracker (Cert 1 has no study log yet)
- [[2026-05-27-claude-certified-architect-foundations-study-plan]] — the parallel Cert 2 escalator
- [[milestones]] — phData onboarding milestones + the $10k cert escalator comp lift
Sources
- Vault:
~/rdco-vault/01-projects/certifications/snowpro-genai-c02/readme.md - Vault:
~/rdco-vault/01-projects/certifications/snowpro-genai-c02/study-plan.md - Vault:
~/rdco-vault/01-projects/certifications/snowpro-genai-c02/study-guide-summary.md - Vault:
~/rdco-vault/01-projects/certifications/snowpro-genai-c02/third-party-resources.md - Vault:
~/rdco-vault/01-projects/certifications/snowpro-genai-c02/flashcards-deck.md - Vault:
~/rdco-vault/01-projects/phdata/cert-progress.md - Vault:
~/rdco-vault/01-projects/phdata/2026-05-27-claude-certified-architect-foundations-study-plan.md - Vault:
~/rdco-vault/01-projects/phdata/milestones.md - Web: Eylon Steiner, "SnowPro Specialty: Gen AI Certification Tutorial (Part 1)," Snowflake Builders Blog — https://medium.com/snowflake/%EF%B8%8F-snowpro-specialty-gen-ai-certification-tutorial-part-1-exam-info-and-starting-with-snowflake-3d01c381d8e9
- Web: Kratika Sharma, "SnowPro Specialty: Gen AI (GES-C01) — My Practical Prep Plan (Feb 2026 Update Included)" — https://medium.com/@sharma.kratika128/snowpro-specialty-gen-ai-ges-c01-my-practical-prep-plan-feb-2026-update-included-1461bd1c9767
- Web: Snowflake official cert page — https://learn.snowflake.com/en/certifications/snowpro-GenAI/ (C02 study guide PDF is auth-gated)
- Web: SnowPro Gen AI Quick Facts — https://certificationpractice.com/exam-overviews/snowflake-snowpro-specialty-gen-ai-quick-facts
- Web: GES-C02 voucher (confirms current exam code) — https://www.learnologyworld.net/products/ges-c02-snowpro-specialty-gen-ai-voucher/5224517000004576086