← Back to catalogue

Level 3 · Ph.D

Ph.D Advanced AI

Artificial Intelligence · Live · v0

Doctoral AI — advanced architectures, training at scale, evaluation science, and an original verified contribution.

Path type

Artificial Intelligence

Requirements

MSc Advanced AI

Enrollment

€19.95

Deadline

Next cohort starts 2027-09-06 (AY 2027/28)

Length

+1 year · 63 phases

Language

English · Español

Content freshness

Curriculum updated 2026-08-22 · content rev 4b78e107

Cortex Credits (CC)

280 CC · 421 CC across all tiers of this programme
What are Cortex Credits?

Syllabus

View the phase syllabus

0. From builder to architect
1. Transformer & attention (architect-depth recap)
2. Modern LLM landscape
3. Decoding & sampling
4. Tokenization & context windows
5. Embeddings & vector spaces
6. Evaluation foundations
7. Prompt anatomy & instruction design
8. Reasoning I: chain-of-thought, decomposition, least-to-most
9. Reasoning II: self-consistency, reflection & debate
10. Tool-augmented reasoning: ReAct & plan-execute
11. Structured & constrained outputs
12. Hallucination I: taxonomy, root causes, when models fabricate
13. Hallucination II: mitigation
14. Prompt optimization
15. Prompt security
16. Context engineering
17. RAG fundamentals
18. Advanced retrieval
19. Advanced RAG: multi-hop, graph RAG, agentic retrieval
20. RAG evaluation
21. Memory systems
22. Feature context & context-as-a-product
23. Knowledge management: corpora, freshness, provenance, governance
24. Tool design: interfaces, affordances, error handling, safety
25. The agent loop: plan–act–observe; single-agent patterns
26. Specialized agents & roles; agent design patterns
27. Multi-agent orchestration I: supervisor/worker, hand-offs
28. Multi-agent orchestration II: parallel fan-out, deterministic workflows, pipelines
29. Skills & modularized AI: reusable capabilities, composition
30. Agent memory, state & long-running/durable tasks
31. Agent reliability: guardrails, validation, human-in-the-loop, recovery
32. Agent evaluation, observability, tracing & debugging
33. Architecture principles: modularization & separation of concerns
34. Blueprints: designing & documenting AI systems for big/complex projects
35. Decomposition & context boundaries; interfaces between AI modules
36. Architectural patterns: router, pipeline, cascade, ensemble, blackboard, fallback
37. Reliability & resilience: retries, circuit breakers, graceful degradation
38. Cost/latency/quality optimization: model routing, caching, distillation choices
39. Security & privacy architecture: PII, secrets, tenant isolation, system-level injection defense
40. Scaling complex AI projects: organizing many agents, skills & contexts; system governance
41. Data strategy for applied AI; pipelines; synthetic data
42. Fine-tune vs RAG vs prompt: decision framework; PEFT/LoRA
43. Evaluation at scale: offline/online, LLM-as-judge, golden sets, regression suites
44. LLMOps I: versioning (prompts/models), registries, CI/CD for AI
45. LLMOps II: monitoring, drift, observability & incident response
46. Productionizing: serving, scaling, cost governance & SLAs
47. AI strategy & value: use-case discovery, ROI, feasibility, build vs buy
48. Applied AI in IT/software: codegen, devops, support, internal knowledge
49. Applied AI across industries I: finance, healthcare, legal (constraints + compliance)
50. Applied AI across industries II: manufacturing, retail, education, public sector
51. Responsible AI & governance: risk, model cards, regulation
52. The AI consultant's craft: discovery, requirements, stakeholders, deliverables, roadmaps, change management
53. Capstone I: discovery → architecture blueprint → modular multi-agent design → eval harness
54. Capstone II: build & deliver — production hardening + consulting deliverable package
55. Frontier & what's next: agentic systems, multimodal, on-device, research signposts; portal integration & program close
56. Serving & deploying inference: batching, caching, autoscaling
57. SLOs, SLIs & error budgets for a probabilistic system
58. Observability & tracing for LLM & agent systems
59. Reliability patterns around model calls: retries, breakers, fallbacks
60. Eval-regression & release gating: CI/CD for models & prompts
61. Drift & quality monitoring + RAG pipeline ops
62. LLMOps/AgentOps & responsible-AI ops: the strand capstone

Syllabus →

Enrolment prerequisites

• A verified account and admissions-committee approval.
• Completion of MSc Advanced AI.