Agenda — Day 5 Part 2: Mock Interview and Portfolio Review¶
The final session converts two weeks of learning into something you can use in a job search: a polished GitHub portfolio, an updated resume, and a rehearsed answer set for the interviews you'll face in the next few weeks.
Session goals¶
By the end of this session you will have:
- A GitHub repo with a README that showcases your capstone project
- A resume bullet that quantifies what you built
- Practiced answers to the 10 most common LLM engineering interview questions
- Completed at least one system design mock with feedback
Schedule¶
| Time | Activity |
|---|---|
| 0:00–0:20 | Resume review and bullet writing |
| 0:20–0:50 | GitHub portfolio polish |
| 0:50–1:30 | Technical Q&A speed round (open book → closed book) |
| 1:30–2:30 | System design mock (pairs) |
| 2:30–3:30 | Full mock interview (one interviewer, one candidate) |
| 3:30–4:00 | Debrief + individual feedback |
What LLM engineering roles actually look for¶
Most job postings list 20 skills. In interviews, they test 5:
- Can you make an LLM do a specific thing reliably? (prompt engineering, structured output, tool use)
- Can you build a production API around it? (FastAPI, async, error handling, deployment)
- Can you make it fast and cheap? (caching, batching, model selection, token budgets)
- Can you evaluate whether it works? (metrics, RAGAS, human eval, A/B testing)
- Can you explain tradeoffs? (RAG vs fine-tuning, LangGraph vs simple chains, streaming vs batch)
Every mock interview question maps to one of these five areas.