14 roles · 166 skills
Every role below is backed by a real list of skills, real labs and real courses, not a wishlist. Pick the one you are aiming at and the plan is built from what you can already do.
Start with the free assessment →STEP 1
Twelve scenario questions, each picked by how you answered the last one. It is free and it writes nothing to a server. The result is a position on every skill it touched, not a score out of ten.
STEP 2
A number of days, up to 180. The plan is built backwards from that date, so a two-week run and a six-month run are genuinely different plans rather than the same list at a different pace.
STEP 3
Mock rounds, practical labs, e-course reading and course evaluations, ordered by what will move your weakest skill soonest and revisited on a spacing schedule rather than once.
STEP 4
Every finished item is evidence. Miss something twice and it moves earlier and comes back more often; pass it and the plan stops spending your days on it.
The counts are what the library actually holds today. Where a role shows fewer labs, the plan leans harder on rounds and course evaluations rather than padding the list.
17 skills · 10 labs · 5 e-courses
2 beginner · 9 intermediate · 6 advanced
Starts with: API contracts, Determinism and temperature, Retries and backoff, Chunking strategy, Grounding answers in sources, Citation correctness.
Ends at: Failure recovery, Agent evaluation harness, Prompt injection, Multi-agent coordination.
45 skills · 6 labs · 2 e-courses
5 beginner · 27 intermediate · 13 advanced
Starts with: Double counting in aggregates, Null semantics, Group-by traps, Inner, outer and anti joins, Hashing and maps, Join fan-out and double counting.
Ends at: Exactly-once versus at-least-once, Clock skew and ordering, Backpressure, Locking and contention.
23 skills · 5 labs · 3 e-courses
1 beginner · 13 intermediate · 9 advanced
Starts with: Sorting, Chunking strategy, Embedding model choice, Index freshness and staleness, Feature pipelines, Choosing an evaluation metric.
Ends at: Data leakage, Train/serve skew, Model and data drift, Inference latency and batching.
82 skills · 6 labs · 2 e-courses
12 beginner · 45 intermediate · 25 advanced
Starts with: Null semantics, Group-by traps, Inner, outer and anti joins, N+1 query patterns, Structured logging, API contracts.
Ends at: Graphs, Greedy strategies, Backtracking, Divide and conquer.
21 skills · 12 labs · 3 e-courses
2 beginner · 15 intermediate · 4 advanced
Starts with: Structured logging, Secret handling, Cost attribution, Autoscaling behaviour, Metric definition, Misleading metrics and averages.
Ends at: DORA metric correctness, Label cardinality explosion, Dependency and supply chain, Test isolation and flakiness.
16 skills · 5 labs · 2 e-courses
5 beginner · 9 intermediate · 2 advanced
Starts with: Double counting in aggregates, Null semantics, Group-by traps, Inner, outer and anti joins, HAVING versus WHERE, Window functions.
Ends at: Data leakage, Model and data drift.
20 skills · 6 labs · 2 e-courses
6 beginner · 13 intermediate · 1 advanced
Starts with: Double counting in aggregates, Null semantics, Group-by traps, Inner, outer and anti joins, HAVING versus WHERE, N+1 query patterns.
Ends at: Reading a query plan.
43 skills · 11 labs · 3 e-courses
3 beginner · 27 intermediate · 13 advanced
Starts with: Structured logging, Secret handling, Sorting, Cost attribution, Autoscaling behaviour, Metric definition.
Ends at: Model and data drift, Inference latency and batching, Dependency and supply chain, Test isolation and flakiness.
45 skills · 12 labs · 3 e-courses
2 beginner · 28 intermediate · 15 advanced
Starts with: Structured logging, Secret handling, Idempotency and replay safety, Cost attribution, Caching strategy, Autoscaling behaviour.
Ends at: Timeout and deadline propagation, Dependency and supply chain, Resource cleanup and leaks, Test isolation and flakiness.
17 skills · 10 labs · 4 e-courses
2 beginner · 9 intermediate · 6 advanced
Starts with: API contracts, Determinism and temperature, Retries and backoff, Chunking strategy, Grounding answers in sources, Citation correctness.
Ends at: Failure recovery, Agent evaluation harness, Prompt injection, Multi-agent coordination.
16 skills · 10 labs · 4 e-courses
1 beginner · 8 intermediate · 7 advanced
Starts with: Determinism and temperature, Chunking strategy, Grounding answers in sources, Citation correctness, Embedding model choice, Context window budgeting.
Ends at: Reranking, Agent evaluation harness, Prompt injection, Inference latency and batching.
22 skills · 5 labs · 2 e-courses
1 beginner · 14 intermediate · 7 advanced
Starts with: Secret handling, Cost attribution, Caching strategy, Autoscaling behaviour, Networking and timeouts, Versioning and compatibility.
Ends at: Split brain and fencing, Data residency, Tenant isolation, Migration strategy.
20 skills · 5 labs · 2 e-courses
0 beginner · 9 intermediate · 11 advanced
Starts with: Join fan-out and double counting, Key stability across loads, Referential integrity without constraints, Cost attribution, Batch versus stream cost, File and storage format choice.
Ends at: Data residency, Read and write path separation, Choosing async boundaries, Migration strategy.
16 skills · 5 labs · 2 e-courses
0 beginner · 11 intermediate · 5 advanced
Starts with: Cost attribution, Autoscaling behaviour, Metric definition, SLOs and SLIs, Alert design and noise, Audit trails.
Ends at: Distributed tracing, Error budgets, Dependency and supply chain, Migration strategy.
The assessment is free and takes about ten minutes. It is what turns the list above into a dated plan built around the things you actually got wrong.
Take the free assessment →