AI / ML Engineer

Turn models into reliable AI features by combining machine learning, software engineering and production operations.

In short— Si ku-call API tu—utajua model, data, evals na vile feature inaship proper.

Free roadmap

What you'll learn

Click any stage kufungua full details.

01Stage 01Engineering basePython, typing and testing • SQL, APIs and data pipelines

Build the coding and data foundation required for reliable AI applications.

Focus on

  • Python, typing and testing
  • SQL, APIs and data pipelines
  • Git, Docker and cloud basics
Build check: Build a tested data-processing API.

Ukimaliza hapa, usirush next stage—build kitu kwanza.

02Stage 02ML foundationsRegression and classification • Feature engineering and validation

Understand training, evaluation and the trade-offs behind model selection.

Focus on

  • Regression and classification
  • Feature engineering and validation
  • Metrics, bias and error analysis
Build check: Train and compare baseline models.

Ukimaliza hapa, usirush next stage—build kitu kwanza.

03Stage 03Deep learningNeural networks and optimization • Transformers and embeddings

Learn the architectures used for modern language, vision and audio systems.

Focus on

  • Neural networks and optimization
  • Transformers and embeddings
  • GPU training and fine-tuning basics
Build check: Fine-tune or adapt a small open model.

Ukimaliza hapa, usirush next stage—build kitu kwanza.

04Stage 04LLM applicationsPrompt and context engineering • RAG, vector databases and reranking

Build grounded and controlled generative-AI workflows.

Focus on

  • Prompt and context engineering
  • RAG, vector databases and reranking
  • Tools, agents and structured output
Build check: Build a cited RAG assistant.

Ukimaliza hapa, usirush next stage—build kitu kwanza.

05Stage 05Evaluation and safetyDeterministic and model-based evals • Guardrails, privacy and prompt injection

Measure quality, manage risk and protect users from bad outputs.

Focus on

  • Deterministic and model-based evals
  • Guardrails, privacy and prompt injection
  • Tracing, latency and cost monitoring
Build check: Create an automated evaluation suite.

Ukimaliza hapa, usirush next stage—build kitu kwanza.

06Stage 06Production AIModel serving and orchestration • Caching, fallbacks and observability

Serve, monitor and improve AI systems as real product features.

Focus on

  • Model serving and orchestration
  • Caching, fallbacks and observability
  • MLOps and continuous evaluation
Build check: Deploy an AI feature with monitoring and fallbacks.

Ukimaliza hapa, usirush next stage—build kitu kwanza.

Now build kitu real.

Pick one project that solves an actual problem, ship it, then explain what you learned. Hapo ndio portfolio inaanza kuhit different.

Choose another path