That scene is why analytics engineering exists. Instead of fixing the dashboard after it breaks, you build the layer underneath — clean, tested, version-controlled SQL models — so a duplicated load fails a test at 6 AM instead of embarrassing someone at 9:47. If you write SQL comfortably and you're tired of being blamed for upstream problems, this is your bridge.
"Analytics engineer" gets about 480 searches a month in India and 1,900 in the US. It's a smaller, more specialised title than "data analyst", and more common at product companies, startups and global capability centres than at services firms (our observation).
Upstream Trace
A dashboard tile shows a wrong number. Tap it, and the page walks you upstream one hop at a time — chart → mart table → staging model → raw file — like following a leak back to the tap. At the raw file you find the cause: duplicate rows. Then the page replays the same morning with one unique test in place: the pipeline stops at 6 AM with a clear message, and the dashboard never shows the wrong number. It's an illustrated demo; no real data is used.
Follow the leak upstream
The skill map
Left: what you have.
| You already have | It becomes | Add this |
|---|---|---|
| SQL queries saved in a folder | dbt models — one SQL file per table, layered | ref() to link models |
| "I always check for duplicates" | Tests — unique, not_null, relationships, accepted_values |
Writing tests in YAML |
| Comments in a shared doc | Documentation — generated docs site | Column descriptions |
| "final_v3_REAL.xlsx" | Version control | Git basics, pull requests |
| Knowing which table finance trusts | Marts — business-ready tables | Naming conventions |
| Explaining numbers to managers | Semantic clarity for BI and AI tools | Metric definitions |
Every right-hand column item is learnable in weeks from a SQL base. Learn time is roughly 5–8 weeks part-time, by our estimate.
Your dbt progress checklist
Tick each step once you've done it on real practice data. Ticks stay on your device.
Most analysts finish the first two in week one.
The tools, checked 24 Oct 2026
- dbt Core — open source under Apache-2.0 (GitHub). Runs from your laptop.
- dbt platform Developer plan — listed as free: one developer seat, 3,000 successful models built per month, one project (dbt pricing). Useful if installing locally is a hassle.
- A warehouse to practise in — the BigQuery sandbox works without a credit card; tables expire after 60 days (docs). PostgreSQL on your laptop also works.
- Git and GitHub — free accounts for public repositories.
- A dashboard on top — Metabase is open source (GitHub), or use Data Studio (formerly Looker Studio).
The proof repo
Build one public GitHub repo:
- Raw: three or four public e-commerce CSVs — orders, customers, products, payments.
- Staging: one model per source, renaming columns and fixing types.
- Marts:
fct_orders,dim_customers, and a monthly revenue table. - Tests: 20 or more. Unique keys, not-null fields, valid relationships, accepted status values.
- Docs: generated docs with column descriptions.
- One dashboard reading only from your marts.
- README headline: "3 data-quality bugs my tests caught." Plant nothing — find real issues in the public data, or document honestly if your tests passed and what they'd catch.
That README line tells an interviewer more than a certificate.
Fourteen days, SQL to dbt
- Days 1–2: Git basics. Make a repo, commit, push.
- Days 3–4: Install dbt Core or open the free developer plan. Connect to your warehouse.
- Days 5–7: Staging models for each source.
- Days 8–10: Marts and the first 10 tests.
- Days 11–12: Ten more tests and column descriptions.
- Days 13–14: Dashboard, README, and five applications.
Your 30/60/90 quests
XP: learning 20 · proof 60 · outreach 40. Ranks: Wayfinder → Builder → Pathmaker. Progress stays in your browser only.
What a normal week looks like
Most analytics engineers split their time between three kinds of work (our observation, not a survey). Building: new models when the business asks a new question. Guarding: reading test failures each morning and deciding whether the data or the test is wrong. Explaining: helping analysts and AI tool owners understand which table to use. If you've ever been the person colleagues asked "which sheet is correct?", you've already done the third part.
Where these roles appear
Search titles: Analytics Engineer, BI Engineer, Data Analyst (dbt), Junior Data Engineer. Check LinkedIn, Naukri, Instahyre and Cutshort, and the careers pages of product companies and global capability centres. We haven't verified pay data for analytics engineering roles; compare live listings and talk to people in the role.
Your realistic 90 days
You wrote SQL queries and cleaned the same data again every week. Analytics engineers turn that cleaning into tested, version-controlled models other analysts reuse. You'll convert queries you've written into dbt models.
Assumes 2–3 hours a day alongside job applications, a laptop and a phone, and free plans only. Every step has a smaller version and a way to pause. Use them. It runs in three blocks of 30 days, each ending with a signal that you're on track. A signal is not a promise of a job or income.
Your first 7 days
| Day | Task | Done when |
|---|---|---|
| 1 | Write down 5 queries you ran repeatedly (logic only, no company data) | 5 queries |
| 2 | Set up dbt Core with DuckDB locally (free) and load a public dataset | dbt runs |
| 3 | Turn 2 queries into staging models | Models build |
| 4 | Add tests | Tests pass |
| 5 | Build one mart model that a dashboard would use | Mart built |
| 6 | Generate dbt docs | Docs site locally |
| 7 | Save 10 analytics engineer postings | 10 links |
Days 1–30: Your queries as a tested project
| Step | Time | Done when | Smaller version | If you need to pause |
|---|---|---|---|---|
| Publish a dbt project: staging, marts, tests and docs | about 14 hours over 3 weeks | repo + docs screenshot | Staging + tests. | Save or commit what works now and stop adding. A finished small version beats an unfinished big one. |
| Share with 5 data people | about 2 hours | 5 messages | Share with 2 data people. | Optional. If reaching out feels like too much this week, skip it and put the time into your proof. |
Signal by day 30: A data engineer reviews your repo.
Days 31–60: Orchestration and BI
| Step | Time | Done when | Smaller version | If you need to pause |
|---|---|---|---|---|
| Schedule runs with GitHub Actions and connect a Looker Studio report | about 10 hours | nightly run + report | Report only. | Stop at the end of the step you're on and write two lines about where you stopped. Pick it up from those notes later. |
| Apply to 8 analytics engineer roles | about 4 hours | 8 logged | Apply to 4 roles. | Stop sending for now. Keep a list of who you've contacted so you can follow up later without doubling up. |
Signal by day 60: A screening call.
Days 61–90: Interviews for analytics engineer roles
| Step | Time | Done when | Smaller version | If you need to pause |
|---|---|---|---|---|
| Practise 10 SQL interview questions (window functions, CTEs) | about 4 hours | 10 done | Practise 5 SQL interview questions (window functions, CTEs). | Bookmark where you stopped. It will still be there next week. |
| Post your project and ask 2 referrals | about 2 hours | post + asks | Asks only. | Optional. If reaching out feels like too much this week, skip it and put the time into your proof. |
Signal by day 90: A technical SQL round.
Job-hunt moves that matter for this route
- Search 'analytics engineer', 'dbt', 'data modeller' on LinkedIn and Instahyre.
- Product companies hire analytics engineers; ex-colleagues there can refer.
- Tests and docs are the proof.
One search that fits your background
The full set of search strings for this move is on the route page. They work on LinkedIn, Naukri and foundit; Instahyre and Cutshort are worth a weekly look too. Add your old job title to any of them with AND, so recruiters who want both skills find you first.
The two worries we hear most about this move
Is analytics engineering just data engineering with a new name? It sits between the two. You model and test data in SQL (often with dbt) so analysts and dashboards can trust it. You won't usually run heavy infrastructure. If you write SQL and care when numbers don't match, you're close already. I've never used Git. Will that hold me back? For a few weeks. Version control is part of the job, but the basics take days. Put your dbt project on GitHub from day one so you learn it while you build.
Other worries, such as age, a career gap, no degree, English, family pressure and savings, are covered on the route page and the role page.
"Data engineering bootcamp" traps
Be careful with courses that promise placement in exchange for upfront fees or ask you to sign loan papers on a call. Never share OTPs or UPI PINs. Official domains only. Report in India at cybercrime.gov.in or 1930; in the US at ReportFraud.ftc.gov.
Analysts moving to dbt ask
Do I need Python? Not to start. SQL and Git are the core. Python helps later.
Snowflake or BigQuery? Either. The dbt ideas are the same; pick the one with an easy free option.
Is this just data engineering? It overlaps, but analytics engineering focuses on business-ready tables and tests rather than infrastructure.
Take the quiz · Back to Data analyst · Route: Analytics engineer · Also: AI-assisted BI
Quick quiz: does this path fit?
Answer each, then press Check. Every answer explains itself.
Make your Path Card
A shareable reminder of where you're heading. The day count starts the first time you open this page; quest XP comes from your log above.
Scam check before you apply anywhere
- Real job and gig platforms are free to join. Never pay a "joining" or "training" fee.
- Never share an OTP or UPI PIN — no genuine employer needs them.
- Use official domains only; type the address yourself.
- India: report at cybercrime.gov.in or call 1930.
- US: report at ReportFraud.ftc.gov.
Small steps, every day.
Each quest has a smaller version for hard weeks, and a pause button. Nothing is lost.
Keep exploring
Sources
Checked 24 Oct 2026. We opened every link below ourselves. Search volumes on this site come from OpenSEO keyword research (Oct 2026).
The other facts here are shared with the route page, where each is listed in full with its link.
| What we checked | Where |
|---|---|
| Metabase open source (AGPL-3.0) | github.com |