AI

AI Master Roadmap

From math foundations to LLMs and MLOps — phased learning path with projects and interview milestones.

Interview tip Pick ONE phase to go deep this month. Ship a project per phase — hiring managers care about demos, not certificates.

① Roadmap overview — five phases

AI learning path
Math & StatsClassical MLDeep LearningLLMs & AppsMLOps
PhaseDurationOutcomeProject
0 — Foundations4–6 weeksLinear algebra, probability, PythonNumPy matrix ops + EDA notebook
1 — Classical ML6–8 weeksSupervised/unsupervised, sklearnKaggle tabular competition top 50%
2 — Deep Learning8–10 weeksCNNs, RNNs, training loopsImage classifier or seq2seq model
3 — LLMs & GenAI6–8 weeksTransformers, RAG, agents, fine-tuningRAG chatbot over your docs
4 — MLOps4–6 weeksDeploy, monitor, CI/CD for modelsAPI endpoint + drift monitoring

② Phase 0 — Math & programming foundations

  • Linear algebra: vectors, matrices, dot product, eigenvalues (PCA intuition)
  • Calculus: derivatives, chain rule (backprop intuition)
  • Probability: Bayes, distributions, expectation, variance
  • Statistics: hypothesis testing, confidence intervals
  • Python: NumPy, pandas, matplotlib; comfortable with Jupyter
  • Git basics: commit, branch, push — all projects in GitHub
Resources: 3Blue1Brown (Essence of LA), StatQuest (YouTube), Khan Academy, "Mathematics for ML" (Deisenroth). Don't get stuck — 80% intuition, move on when you can follow ML lectures.

③ Phase 1 — Classical machine learning

TopicKey conceptsInterview must-know
SupervisedRegression, classification, train/val/test splitBias-variance, cross-validation
AlgorithmsLinear/logistic regression, trees, RF, XGBoost, SVMWhen to use which; interpretability
UnsupervisedK-means, PCA, anomaly detectionChoosing k; PCA for dimensionality reduction
EvaluationAccuracy, precision, recall, F1, ROC-AUCImbalanced classes → do not trust accuracy
Feature engScaling, encoding, missing values, leakageTarget leakage is a common interview trap
Project: End-to-end sklearn pipeline on a real dataset — document EDA, feature choices, and why you picked the final model. Deploy nothing yet; focus on methodology.

④ Phase 2 — Deep learning

Neural network basics — layers, activations, loss, optimizer
Backpropagation intuition — chain rule through the graph
CNNs — conv, pool, ResNet; image classification
RNNs/LSTMs — sequences, vanishing gradient; time-series/NLP pre-transformer
Training — batch size, learning rate schedule, regularization (dropout, weight decay)
Frameworks — PyTorch preferred for research/jobs; TensorFlow for some enterprise
Resources: fast.ai Practical Deep Learning, Andrew Ng Deep Learning Specialization, PyTorch tutorials. Project: Train a CNN on CIFAR-10 or fine-tune a small model — log metrics in TensorBoard or W&B.

⑤ Phase 3 — LLMs & generative AI

TopicWhat to learnHands-on
TransformersSelf-attention, positional encoding, encoder-decoderRead "Attention Is All You Need" summary
Pre-trained modelsGPT, BERT, T5 — pretrain vs fine-tune vs promptHugging Face model hub experiments
Prompt engineeringSystem prompts, few-shot, chain-of-thoughtBuild eval set of 20 prompts
RAGEmbeddings, chunking, vector DB, retrieval + generationChatbot over PDF/wiki
Fine-tuningLoRA, RLHF concept, when to fine-tune vs RAGFine-tune small model on domain data
AgentsTool use, ReAct loop, guardrailsAgent with 2 tools (search + calculator)
Use cheat sheets on this site: RAG, First LLM Integration, Agents Intro, Prompt Engineering.

⑥ Phase 4 — MLOps & production

  • Model serving: FastAPI + ONNX/TorchServe/vLLM for LLMs
  • Containerization: Docker basics; deploy to AWS/GCP/Fly.io
  • Monitoring: latency, error rate, data drift, model drift
  • CI/CD: retrain pipeline triggered on new data
  • Cost: token usage, GPU hours, cache embeddings
  • Security: API keys in secrets manager, input validation, PII in logs
Minimal ML production stack
Client / App
API Gateway + Auth
Model Server (GPU/CPU)
Vector DB + Feature Store
Logging / Metrics (Prometheus, Grafana)

⑦ Role-specific tracks

Target roleEmphasizeDe-emphasize
ML EngineerTraining pipelines, distributed training, model optimizationHeavy frontend
Data ScientistEDA, experimentation, causal inference, storytellingCUDA kernel tuning
AI App DeveloperLLM APIs, RAG, agents, product UXProof of convergence theorems
Research ScientistPaper reading, novel architectures, benchmarksCRUD app building
Most industry roles in 2025–2026 are AI application + ML engineering hybrid — strong Python, LLM integration, and evaluation beat pure theory.

⑧ 12-week sprint plan (template)

WeeksFocusDeliverable
1–2Math refresh + PythonGitHub repo with NumPy exercises
3–4Classical ML + KaggleOne competition submission + write-up
5–6PyTorch + CNN projectImage model with training curves logged
7–8Transformers + Hugging FaceFine-tune or zero-shot eval notebook
9–10RAG applicationDeployed chatbot (even on free tier)
11Mock ML interviews5 timed Q&A sessions recorded
12Portfolio polishREADME, demo video, LinkedIn post
Adjust pace: 10–15 hrs/week realistic alongside a job. Double timeline if <8 hrs/week.

⑨ Interview prep milestones

  • Explain bias-variance without notes in under 2 minutes
  • Whiteboard: train/val/test split and k-fold CV
  • Describe attention mechanism at high level
  • Compare RAG vs fine-tuning with a concrete example
  • Walk through one project end-to-end (problem → data → model → metric → result)
  • Answer "what would you do with imbalanced classes?" with 3 techniques
  • Discuss one failure: wrong metric, data leakage, or overfitting you fixed
roadmapMLdeep-learningLLMMLOpscareer

⑩ Revision checklist

  • Identified current phase and next project
  • Balanced theory with shipped projects on GitHub
  • Know which role track to optimize for
  • Have 12-week plan with weekly deliverables
  • Prepared 2-minute project stories for interviews
  • Connected roadmap to site cheat sheets (RAG, agents, etc.)