AI/ML Engineering Foundation
Set up your professional AI engineering environment, master LangChain LCEL, prompt engineering patterns, and end-to-end tracing with LangSmith.
Explore Kit 01
Modular, project-driven AI engineering programs designed for working professionals. Pick your specialization, work through hands-on labs, and graduate kit-credentialed into production AI.
Every kit is designed around a real deliverable — not a slide deck. You finish with running code, a portfolio project, and the confidence to ship AI in production.
Pick the kits that match your current role or career goal. Do one kit or complete the full specialization path — the program flexes around you.
Labs come before lectures. Each phase starts with a hands-on exercise that reveals why the theory matters, not the other way around.
Complete a kit, submit your capstone, earn a verifiable kit credential. Collect all five to earn the Applied AI Engineering certification.
Work with PostgreSQL, MongoDB, Redis, and Pinecone — not toy datasets. Connect to live APIs and build pipelines that handle real data at scale.
Get code reviews and architecture feedback from SelectAI engineers. Ask questions, share your work, and unblock yourself faster.
Every capstone includes a deployment milestone — containerized, cloud-hosted, observable. You'll know your code runs because it does run.
Each kit is a self-contained learning module with labs, a mentor review, and a capstone project. Browse them on the learning platform.
Set up your professional AI engineering environment, master LangChain LCEL, prompt engineering patterns, and end-to-end tracing with LangSmith.
Explore Kit 01Connect AI to real databases — PostgreSQL, MongoDB, Redis. Build retrieval-augmented generation pipelines with FAISS and Chroma vector stores.
Explore Kit 02Build stateful, branching, and parallel agent workflows with LangGraph. Implement MCP servers and coordinate multi-agent pipelines that solve real problems.
Explore Kit 03Red-team your own AI systems. Implement input/output guardrails, bias audits, content moderation pipelines, and enterprise AI governance frameworks.
Explore Kit 04Fine-tune open-source LLMs with LoRA/QLoRA, run models locally with Ollama, containerize with Docker, and deploy observable AI services to the cloud.
Explore Kit 05The full 20-week path across 9 phases, organized across all 5 kits. Each phase has a concrete build objective — you always know what you're working toward.
Professional Python setup, virtual environments, VS Code for AI development, Git discipline, Groq API, and first prompt traces in LangSmith.
Advanced prompt patterns, chain-of-thought, few-shot learning. Build composable chains with LangChain Expression Language (LCEL).
Design schemas optimized for AI workloads. Connect LangChain to PostgreSQL, MongoDB, and Redis. Handle structured and unstructured data in the same pipeline.
Build end-to-end RAG pipelines with FAISS, Chroma, and Pinecone. Implement hybrid search, reranking, and evaluation frameworks.
Build ReAct agents with custom tools, structured output, and error recovery. Implement human-in-the-loop checkpoints for high-stakes decisions.
Design stateful, branching multi-agent graphs with LangGraph. Build and expose custom Model Context Protocol (MCP) servers. Coordinate parallel agent sub-graphs.
Implement prompt injection defenses, jailbreak testing, input/output filters, PII detection, and content moderation pipelines.
Fine-tune Llama 3 and Mistral with LoRA and QLoRA. Run models locally with Ollama. Compare fine-tuned vs base models with structured evals.
Containerize AI services with Docker, build CI/CD pipelines with GitHub Actions, add observability with LangSmith and Prometheus, and deploy to AWS or GCP.
Every lab starts with a working failure — you fix it, extend it, and ship it. No copy-paste exercises.
Build a production RAG system with hybrid retrieval, metadata filtering, and cross-encoder reranking. Evaluate with RAGAs.
8 labs in Kit 02Design a customer support agent with branching logic, tool fallbacks, and a human escalation checkpoint using LangGraph state machines.
10 labs in Kit 03Run structured attacks against your own LLM app, document vulnerabilities, and implement defenses using Guardrails AI and NeMo Guardrails.
8 labs in Kit 04Fine-tune Llama 3.1 on a domain-specific dataset with QLoRA, run the model locally via Ollama, and benchmark latency vs hosted APIs.
9 labs in Kit 05Wire PostgreSQL, MongoDB, and a vector store into a single LangChain retrieval chain with caching via Redis and full observability.
9 labs in Kit 02Containerize your final capstone project, set up a GitHub Actions CI/CD pipeline, and deploy to AWS ECS with health checks and alerting.
9 labs in Kit 05Access a single kit or the full specialization path. All plans include the learning platform, labs, and kit credential on completion.
Single Kit
Perfect if you have a specific gap to fill. One kit, all labs, kit credential on completion.
Full Specialization
The complete Applied AI Engineering path — all 5 kits, 4 capstones, and the full specialization credential.
Enterprise / Teams
Bring Applied AI Engineering to your team. Custom cohort scheduling, dedicated mentor, and branded learning experience.