Role roadmaps / AI Engineer Role roadmap
Ship products on pretrained models: retrieval, eval, and guardrails. The hard parts are data, not the model brand.
Also called: Applied AI Engineer, LLM Application Engineer
You ship: A retrieval pipeline with evals, access control, and a rollback — not a chat demo.
Faster to demo, slower to production than people admit
Roadmap Coding Interview Question bank Switch roleData Analyst BI Analyst Data Engineer Analytics Engineer Data Scientist ML Engineer MLOps Engineer AI Engineer Data Architect Nearby roles:ML Engineer , Data Engineer , Data Scientist
→
Skill check10 questions Already working in this role? Answer honestly and the roadmap ticks what you have done, then points you at the first gap.
Take the 2-minute skill check 1 of 10
Can you do this?
1 Yes, I have done this2 Partly3 Not yet
Back Keys 1, 2 and 3 answer while the check has focus.
Result
Ticks are only added, never removed. Partly and Not yet leave a skill unticked.
Checked skills
0% · pick any skill to see its syllabus, links and practice.
Journey Map List
− Fit + Print Clear checks
RTE
Stop 1 of 16
◀ Prev StopNext Stop ▶ Auto Cruise
Click any bus stop to drive there and inspect the complete syllabus, or scroll the highway to follow the bus.
Terminal Start · Station 1 of 16
Click to stop Start · AI Engineer A retrieval pipeline with evals, access control, and a rollback — not a chat demo. Inspect station brief →
Checkpoint · Station 2 of 16
Click to stop Foundations The tools you cannot fake in week one Inspect station brief →
Bus Stop · Station 3 of 16
Click to stop RAG is ETL Chunk, embed, version, ACL-filter. Treat the index like a table. Open Syllabus →
Bus Stop · Station 4 of 16
Click to stop Warehouse-native LLMs COMPLETE in incremental jobs, never in a dashboard refresh. Open Syllabus →
Bus Stop · Station 5 of 16
Click to stop Eval before vibes Golden questions, faithfulness, and a human sample. Latency is a metric. Open Syllabus →
Checkpoint · Station 6 of 16
Click to stop Daily craft What you actually do on a Tuesday Inspect station brief →
Bus Stop · Station 7 of 16
Click to stop Prompts as code Version them. Test them. Do not hide them in Slack. Open Syllabus →
Bus Stop · Station 8 of 16
Click to stop Retrieval quality Hybrid search, metadata filters, and “why this chunk?” Open Syllabus →
Bus Stop · Station 9 of 16
Click to stop Safety and ACL The model must not see rows the user cannot. Filters happen before generation. Open Syllabus →
Checkpoint · Station 10 of 16
Click to stop How you ship Production habits, not extra courses Inspect station brief →
Bus Stop · Station 11 of 16
Click to stop Traces, not screenshots Log retrieval, tokens, and thumbs-down. Cost per answer. Open Syllabus →
Bus Stop · Station 12 of 16
Click to stop Orchestrate like a pipeline Airflow or a job for re-embed. The chatbot is the last mile. Open Syllabus →
Checkpoint · Station 13 of 16
Click to stop Career moves How this role talks in interviews Inspect station brief →
Bus Stop · Station 14 of 16
Click to stop AI engineer interview A RAG design, eval, and “what happens when the source table changes?” Open Syllabus →
Bus Stop · Station 15 of 16
Click to stop Next role MLE if you train. DE if retrieval was the interesting part. Product if you like users. Open Syllabus →
Final Destination · Station 16 of 16
Click to stop Keep practicing SQL + Python for this role Inspect station brief →
Start · AI Engineer Foundations RAG is ETL Chunk, embed, version, ACL-filter. Treat the index like a table. Warehouse-native LLMs COMPLETE in incremental jobs, never in a dashboard refresh. Eval before vibes Golden questions, faithfulness, and a human sample. Latency is a metric. Daily craft Prompts as code Version them. Test them. Do not hide them in Slack. Retrieval quality Hybrid search, metadata filters, and “why this chunk?” Safety and ACL The model must not see rows the user cannot. Filters happen before generation. How you ship Traces, not screenshots Log retrieval, tokens, and thumbs-down. Cost per answer. Orchestrate like a pipeline Airflow or a job for re-embed. The chatbot is the last mile. Career moves AI engineer interview A RAG design, eval, and “what happens when the source table changes?” Next role MLE if you train. DE if retrieval was the interesting part. Product if you like users. Keep practicing Phase 1: Foundations RAG is ETL Chunk, embed, version, ACL-filter. Treat the index like a table. Warehouse-native LLMs COMPLETE in incremental jobs, never in a dashboard refresh. Eval before vibes Golden questions, faithfulness, and a human sample. Latency is a metric. Phase 2: Daily craft Prompts as code Version them. Test them. Do not hide them in Slack. Retrieval quality Hybrid search, metadata filters, and “why this chunk?” Safety and ACL The model must not see rows the user cannot. Filters happen before generation. Phase 3: How you ship Traces, not screenshots Log retrieval, tokens, and thumbs-down. Cost per answer. Orchestrate like a pipeline Airflow or a job for re-embed. The chatbot is the last mile. Phase 4: Career moves AI engineer interview A RAG design, eval, and “what happens when the source table changes?” Next role MLE if you train. DE if retrieval was the interesting part. Product if you like users.
RAG is ETL Chunk, embed, version, ACL-filter. Treat the index like a table.
Checked automatically: you solved the linked practice problems.
RAG is ETL: chunk, embed, version, ACL-filter. Treat the index like a table.
Time to learn Chunk, embed, version, ACL-filter. Index is a table.
Prerequisites Grain of a chunk Who may retrieve it What you need RAG pipelines article AI questions A corpus with an owner Concepts to study Chunking as a modeling choice Embedding model versioning Index refresh as a pipeline ACL filters before retrieval Citations / chunk IDs as grain Tech stacks Warehouse or object store as source — System of recordVector DB or warehouse vectors — IndexOrchestrator — RefreshHow to study RAG pipelines article; AI interview questions. Draw source → chunk table → embed job → index → query with ACL. What to produce A chunk grain sentence An ACL filter before embedding or at query Practice on this site RAG pipelines; AI questions Pitfalls Re-embedding in the request path No ACL so the model sees rows the user cannot Unversioned chunks Interview prompts What is the grain of a chunk, and who is allowed to retrieve it? Links
Previous skill Next skill
Mark as done Search DataLane All Articles Stacks Practice Learn Pages
No results. Try analyst, numpy, snowflake, or roadmap.
↑↓ navigatealt ↑↓ jump group↵ openesc closeAliases: sf, numpy, analyst, mlops