AI & GenAI Engineering
Build intelligent production applications: LLM orchestration, Retrieval-Augmented Generation (RAG), vector databases, and private AI agent workflows.
What You Build
Week-by-Week Technical Sprints
LLM Foundations & Prompt Engineering Architecture (DRAFT)
- •Tokenization, context windows, and latency vs cost trade-offs
- •Few-shot prompting, structured JSON outputs, and function calling
Vector Embeddings & Semantic Search (DRAFT)
- •Embedding models, distance metrics (Cosine, Euclidean), and chunking strategies
- •Vector database integration: schema, indexing, and hybrid BM25 search
Advanced Retrieval-Augmented Generation (RAG) (DRAFT)
- •Parent-child retrieval, re-ranking models, and query expansion
- •Context compression and dynamic window optimization
Agentic Workflows & Tool Execution (DRAFT)
- •ReAct prompting pattern, autonomous multi-step reasoning, and tool execution
- •State machines with LangGraph and loop termination controls
RAG Evaluation, Benchmarking & Guardrails (DRAFT)
- •RAGAS metrics: faithfulness, answer relevancy, and context recall
- •NeMo Guardrails, prompt injection mitigation, and safety filters
Local LLMs & Open-Source Inference (DRAFT)
- •Ollama, vLLM, and HuggingFace model deployment
- •Quantization (GGUF, AWQ) and private self-hosted inference
Full-Stack AI Application Integration (DRAFT)
- •Streaming responses via Vercel AI SDK to Next.js interfaces
- •Handling token limits, backpressure, and persistent chat sessions
Fine-Tuning Foundations & Dataset Curation (DRAFT)
- •When to fine-tune vs when to use RAG
- •LoRA / QLoRA techniques and dataset preparation / deduplication
Capstone Project: Domain AI Copilot (DRAFT)
- •Architecting a specialized AI system for Healthcare or Legal workflows
- •HIPAA/privacy isolation and zero-data retention policies
Hardening, Deployment & Final Review (DRAFT)
- •Cost monitoring, rate-limiting, caching repeated queries, and deployment
- •Final project review and credential issuance
Capstone: Private Clinical Diagnostic & Consultation Summarizer
A secure, privacy-first AI workflow that ingests doctor-patient dialogue, extracts clinical telemetry, cross-references internal medical protocols via RAG, and drafts consultation summaries without exposing patient identifiers.
Capstone Production Deliverables:
- ✓Streaming Next.js clinical dashboard with role-based doctor authentication
- ✓Hybrid semantic search pipeline indexed over medical guideline repositories
- ✓Automated RAG evaluation report documenting accuracy and faithfulness
Be Notified First When Enrollment Opens
AI & GenAI Engineering is currently undergoing curriculum task banking and mentor onboarding. Join the waitlist for priority batch access and launch discounts.
How You Are Assessed
Milestone evaluation across 10 weekly sprints, automated RAG evaluation metrics, code quality, and capstone presentation.
Certificate Eligibility
- •Completion of all core weekly technical deliverables
- •Successful execution of the Capstone AI platform with benchmark report
- •Passing code review with assigned senior AI mentor
24-Hour Refund Policy
Students may request a refund within 24 hours of payment, and it is refundable. To request a refund, submit a request via our refund policy desk.