Resume & Interview Preparation for AI/ML Jobs
🎯 TL;DR — This cheat sheet gives you a complete, copy-paste-ready AI/ML resume, cover letter, LinkedIn headline, cold outreach DM, and job search playbook. Stop starting from scratch — use these templates.
📋 Core Resume Framework
| Section | What to Include | ATS Tip |
|---|---|---|
| Header | Name, email, phone, LinkedIn, GitHub, location (city only) | No tables/columns — plain text |
| Summary | 2–3 lines: role title + top skills + one career achievement | Include job title keyword exactly |
| Skills | Categorized by type (see template below) | Match exact terms from job descriptions |
| Projects | 2–4 AI projects with tech stack + metrics | Use numbers: "reduced latency by 40%" |
| Experience | Reverse chronological, STAR-formatted bullets | Action verb + tool + quantified result |
| Education | Degree, institution, year — certifications below | List relevant courses if recent grad |
1-page rule: For under 3 years experience, keep it to 1 page. Every bullet must earn its spot.
🔢 Step-by-Step: Building Your AI Resume
- Start with the job description — copy/paste it into a doc, highlight every technical skill mentioned. These become your keywords.
- Write your skills section first — map your actual skills to the JD keywords. Don't lie; do use exact terminology.
- Write project bullets — use format: "Built [what] using [tech], achieving [metric]." Add GitHub link.
- Write experience bullets — same format. Start with action verb. Quantify every bullet.
- Write summary last — 2–3 lines that echo the job title and top 3 skills from the JD.
- ATS check — paste your resume into a plain .txt file. If it looks garbled, fix formatting.
- Human check — show to a senior engineer. Ask: "Does this pass the 6-second scan?"
💡 Templates & Scripts
Skills Section (copy, edit to match your stack)
Programming Languages: Python, SQL, R
ML Frameworks: Scikit-learn, PyTorch, TensorFlow, XGBoost
AI/LLM: LangChain, LangGraph, OpenAI API, Anthropic Claude API, HuggingFace Transformers
RAG & Vector DBs: ChromaDB, Pinecone, FAISS, LlamaIndex
MLOps: MLflow, Docker, FastAPI, Streamlit, GitHub Actions
Cloud: AWS SageMaker | GCP Vertex AI | Azure ML
Data: Pandas, NumPy, Matplotlib, Seaborn
Databases: PostgreSQL, MySQL, MongoDBProject Bullet Templates (fill in the brackets)
• Built an end-to-end RAG pipeline using LangChain + ChromaDB + Claude API,
reducing document Q&A response time by 60% vs. keyword search baseline.
• Developed a multi-agent AI system with LangGraph for [task], processing
[X] queries/day with [Y]% accuracy on held-out test set.
• Fine-tuned [model name] on [domain] dataset using LoRA + QLoRA, achieving
[metric] with 4x lower training cost vs. full fine-tuning.
• Deployed [ML model] as REST API using FastAPI + Docker on AWS EC2,
serving [X] predictions/day with <100ms p95 latency.
• Built [NLP application] using HuggingFace Transformers, improving [metric]
from [baseline] to [result] on [dataset].Cover Letter Template (200 words — scannable)
Subject: AI Engineer Application — [Your Name]
Hi [Hiring Manager Name / "Hiring Team"],
I'm applying for the [Job Title] role at [Company].
In my most recent project, I built [Project Name] — [one-sentence description of what it does
and what it achieved]. I used [2–3 key technologies from their JD], which I see are central
to your stack.
What draws me to [Company] specifically: [1–2 genuine, specific reasons — mention a product
feature, research paper, blog post, or company mission statement you actually care about].
I bring:
• [Skill 1 relevant to JD] — used in [specific project]
• [Skill 2 relevant to JD] — demonstrated by [specific result]
• [Skill 3 relevant to JD] — [brief context]
My GitHub with relevant projects: [link]
My portfolio/demo: [link]
Happy to do a quick screen call at your convenience. I'm available [times].
Best,
[Your name] | [email] | [LinkedIn]LinkedIn Headline Formulas
Formula 1 (skill-led):
AI Engineer | LangChain · RAG · Python | Building production LLM applications
Formula 2 (value-led):
Building AI Agents that save teams 10+ hours/week | Python · LangGraph · Claude API
Formula 3 (aspiration, for job seekers):
Aspiring AI Engineer | LLM Apps · RAG Systems · Python | Open to Roles in AI/ML
Formula 4 (achievement-led):
AI Engineer | Deployed 5 AI apps to production | LangChain · FastAPI · AWSCold Outreach DM Template (LinkedIn / Email)
Hi [Name],
I came across your work on [specific project/post/company] — [one genuine specific thing
that caught your eye].
I'm an AI engineer working with [LangChain/RAG/Agents — whatever is relevant].
I recently built [brief project description with link].
I'd love to learn more about [their team/role/work] — would you be open to a 15-minute
chat? No pressure at all if you're busy.
[Your name]
GitHub: [link] | Portfolio: [link]Salary Negotiation Script
"Thank you for the offer — I'm genuinely excited about this role and the team.
Based on my research on [Levels.fyi / LinkedIn Salary / Glassdoor] and the scope of
this position, I was expecting something closer to [target number].
Is there flexibility to get to [target]? I'm very motivated to join and would love
to make this work."
[If they can't move on base]: "I understand. Could we explore [signing bonus /
additional equity / earlier review cycle / remote flexibility]?"🎤 Practice Q&A
Q: Why do you want to work in AI engineering?
A: "I've been drawn to the intersection of software and intelligence — specifically the moment when a model you built does something genuinely useful in the real world. My turning point was building a RAG system for [topic] — it could answer questions better than a keyword search I'd spent weeks on. That experience convinced me this is where I want to spend the next decade."
Q: Walk me through your most impressive AI project.
A: "My most technically challenging project was [name]. The problem: [1 sentence]. I built a pipeline using [tech stack]. The hardest part was [specific challenge] — I solved it by [approach]. The result: [metric]. I can share the GitHub — it has a full README and demo."
Q: How do you stay current with AI advancements?
A: "I follow [Andrej Karpathy, Simon Willison, Hugging Face blog]. I read 1–2 papers per week from Arxiv (I use the abstract + intro + results section method — rarely need the full paper). I also build small experiments with new tools — my GitHub has several weekend projects where I tested new models or frameworks."
⚠️ Common Mistakes to Avoid
- Generic summaries like "passionate data scientist seeking opportunity" — be specific
- No GitHub links — every project bullet needs a link; a resume without GitHub is invisible
- "Worked on" or "helped with" — use strong verbs: built, deployed, optimized, reduced, improved
- Lying about skills — if you list PyTorch, they WILL ask you to implement backprop
- Same resume for every job — swap the summary + reorder skills to match each JD
- Applying without a referral when one is possible — referrals 4x your callback rate
- Following up more than once — one polite follow-up after 7 days is professional; more is not
🚀 Quick Reference: Top Job Boards for AI/ML
| Platform | Best For |
|---|---|
| LinkedIn Jobs | Volume, all company sizes |
| Levels.fyi | Compensation data + jobs at top companies |
| Wellfound (AngelList) | AI startups |
| Y Combinator jobs board | YC-backed AI companies |
| Anthropic Careers | anthropic.com/careers |
| Hugging Face Jobs | AI/ML-focused roles |
| ai-jobs.net | Curated AI-only listings |
Apply-to-referral ratio target: 60% direct apply, 40% via referral or cold outreach.
📋 Action Checklist
- Fill in the Skills Section template with your real stack
- Write 3 project bullets using the template format (verb + tech + metric)
- Write or rewrite your cover letter using the 200-word template
- Update your LinkedIn headline using one of the 4 formulas
- Set up LinkedIn job alerts for "AI Engineer", "ML Engineer", "LLM Engineer"
- Send 5 cold outreach DMs this week using the template above
- Add your resume to levels.fyi and Wellfound profiles
- Practice your salary negotiation script out loud with a friend