Skip to content

Resume Checklist for LLM Engineering Roles

A resume for LLM engineering roles is different from a general software engineering resume: recruiters and hiring managers are looking for specific signals that you have built real systems with real AI components, not just completed tutorials.

Learning objectives

  • Write resume bullets that quantify LLM project work
  • Identify and list the skills that LLM engineering job postings actually screen for
  • Avoid the most common resume mistakes for AI/ML roles

Skills section — what to include

List these only if you can speak to them in an interview. A skill on your resume is an invitation to be asked about it.

Core LLM skills (high signal)

  • OpenAI API, Anthropic API — not just "LLMs"
  • RAG (Retrieval-Augmented Generation) with the specific stack: ChromaDB, Pinecone, Qdrant
  • LangChain, LangGraph — name the version era you know (LCEL, not legacy)
  • Function calling / tool use — OpenAI tools API, Anthropic tool_use
  • Prompt engineering — few-shot, chain-of-thought, structured output
  • Fine-tuning — LoRA, QLoRA, PEFT, transformers, SFTTrainer
  • Streaming — SSE, FastAPI StreamingResponse
  • LLMOps — LangSmith, cost tracking, latency optimization

Deployment and infrastructure (expected)

  • FastAPI, uvicorn, Pydantic
  • Docker, Modal, Fly.io (whichever you've used)
  • async/await, asyncio — important signal for LLM services

Evaluation (differentiator — most candidates skip this)

  • RAGAS — faithfulness, answer relevancy, context precision
  • LangSmith — tracing, datasets, evaluators
  • Metrics: latency P95, cache hit rate, token cost per query

How to write resume bullets for LLM projects

The formula: Action verb + what you built + technology used + measurable outcome

Bad examples

  • "Built a chatbot using LangChain and OpenAI"
  • "Implemented RAG pipeline"
  • "Used LLMs to build a question answering system"

Good examples

  • "Built a RAG customer support bot with FastAPI and ChromaDB that reduced average answer latency from 4.2s to 380ms via semantic caching, handling 200 req/day"
  • "Implemented a LangGraph multi-agent research pipeline with planner → researcher → critic loop; achieved 0.82 average quality score across 20 test queries"
  • "Deployed a document extraction API using OpenAI function calling and Pydantic validation; achieved 84% field-level accuracy on a 15-document test set"
  • "Reduced LLM API costs by 40% using prompt caching and exact-match response caching on deterministic queries"
  • "Designed and evaluated a QLoRA fine-tuning pipeline for sentiment classification; improved F1 from 0.71 (zero-shot) to 0.89 on held-out test set"

What to include for each project

For each project in your portfolio, your resume bullet should answer:

  1. What problem did it solve?
  2. What technologies did it use?
  3. What was a measurable outcome?
✓ "Built a FastAPI RAG service (ChromaDB + gpt-4o-mini) for customer Q&A;
   evaluated 20 test questions, 78% keyword precision, P95 latency 1.8s"

✗ "Created a question answering chatbot"

Common resume mistakes for AI/ML roles

Listing every library you've imported. If you have numpy, pandas, scikit-learn, tensorflow, pytorch, transformers, langchain, openai all in the skills section, it signals you've touched all of them but mastered none. Be selective.

No metrics. "Improved performance" is not a metric. "Reduced P95 latency from 3.1s to 0.9s with a caching layer" is.

Vague project descriptions. "Worked on LLM projects at [Company]" tells a hiring manager nothing. Name the system, name the technology, name the outcome.

Listing GPT-4 experience from the ChatGPT UI. Using the ChatGPT web interface is not the same as building with the OpenAI API. Be accurate.

Hiding evaluation. The fact that you ran a quantitative evaluation is a differentiator. Most candidates only demo. Put the number in the bullet.

One project with three metrics beats three projects with no metrics

Depth signals expertise. A capstone project with measured latency, cache hit rate, and answer quality is more credible than five tutorial projects with no outcomes.

Resume structure for junior/mid LLM engineering roles

[Name] | [Email] | [GitHub] | [LinkedIn]

## Summary (2–3 sentences)
LLM engineer with hands-on experience building RAG pipelines,
LangGraph agents, and production FastAPI services. Comfortable
with the full stack: prompt engineering → API integration →
async deployment → quantitative evaluation.

## Skills
LLM APIs: OpenAI (gpt-4o-mini, embeddings), Anthropic (claude-haiku-4-5)
Frameworks: LangChain (LCEL), LangGraph, PEFT
Infrastructure: FastAPI, Docker, Modal, ChromaDB, Pinecone
Evaluation: RAGAS, LangSmith, custom metrics
Languages: Python (primary), SQL

## Projects
[Capstone] RAG Customer Support Bot ...
[Project 2] LangGraph Research Agent ...

## Experience
...

## Education
...

00-agenda | 02-portfolio-and-github