AI6 months

AI Engineering

Building reliable products on top of models: retrieval, agents, evals and cost control.

  1. Step 01

    Python & APIs

    Python fluency, async HTTP, typed data models and calling model APIs correctly.

    PythonPydantic

  2. Step 02

    LLM fundamentals

    Tokens, context windows, temperature, structured output, function calling and where models fail.

    OpenAIClaudeGemini

  3. Step 03

    Prompt engineering

    System design of prompts, few-shot patterns, decomposition and refusal handling.

    PromptingJSON mode

  4. Step 04

    RAG

    Chunking, embeddings, vector stores, hybrid search, reranking and citation-grounded answers.

    pgvectorEmbeddings

  5. Step 05

    Agents & tools

    Tool schemas, planning loops, guardrails, sandboxing and human-in-the-loop escapes.

    AgentsTool calling

  6. Step 06

    Evaluation

    Golden datasets, LLM-as-judge with care, regression suites and offline vs online metrics.

    EvalsTracing

  7. Step 07

    Production

    Streaming, caching, fallbacks across providers, rate limits, cost dashboards and PII handling.

    LatencyCostSafety