Skip to main content

Demos, risk logs, escalation

Client-oda vaaram-oru rhythm nadathunga: velai seiyra software-oda oru short demo, owners-oda update panna risk log, edhaavadhu block aanaa mudhalla, amaidhiyaa escalation.

Pro 3 · AI system designerabout 2 hours5 sections

Short-aa sonna

Client-a vaaram-oru demo-vum, owners, dates-oda risk log-um, block aanaa mudhalla amaidhiyaa escalation-um vachchu oru step-la vachchukkunga.

Idhu yen mukkiyam

Kettha seidhi late-aa vandhaa nambikkai poidum; plan-oda mudhalla vandhaa nambikkai valarum. Steady-aana vaaram-oru rhythm client team-a engage-aa vachchirukkum, problems-a selavu kammiyaa irukkum podhae veliya kondu varum, neenga kooda velai seithaa client-ku nyabagam irukkradhu idhu thaan.

Sign up free to read the whole lesson

The rest of this lesson: 5 parts, video lessons, the best course, practice and a quick check. It is free, takes a few seconds and needs no card.

Accounts are for learners aged 18 and over. Under 18? The quick answer and the first part of every lesson stay open: browse lessons by topic.

Lessons on Forward Deployed Engineer (FDE)0 of 90 done
  1. 1Python for production
  2. 2Project tooling: environments, tests and logs
  3. 3SQL for engineers
  4. 4REST APIs with FastAPI
  5. 5Files, tables and PDFs
  6. 6Git with a team
  7. 7Linux, the shell and SSH
  8. 8Docker and Compose
  9. 9Call an LLM API from code
  10. 10Tokens and the context window
  11. 11How a model picks its next word: temperature, top-p and top-k
  12. 12AI in Indian languages: tokens, quality gaps and what to use
  13. 13System prompts and prompt design
  14. 14Get structured JSON output
  15. 15Tool calling
  16. 16Embeddings
  17. 17Chunk documents for retrieval
  18. 18Vector search
  19. 19Build a RAG app
  20. 20Check whether retrieval works
  21. 21Hybrid search, reranking and filters
  22. 22Query rewriting, HyDE and parent-child chunks
  23. 23Grounding, citations and refusing to guess
  24. 24Long documents without vectors
  25. 25Test sets and evals
  26. 26LLM judges and their limits
  27. 27Choosing a model: benchmarks, leaderboards and your own tests
  28. 28Scoring open-ended answers: exact checks, similarity and side-by-side comparison
  29. 29Catching hallucinations: why models make things up and how to detect it
  30. 30Regression tests for prompts
  31. 31Quality, cost and speed trade-offs
  32. 32The agent loop
  33. 33Designing tools for agents
  34. 34Agent memory
  35. 35Planning and reflection
  36. 36LangGraph: agents as graphs
  37. 37Agent SDKs and sub-agents
  38. 38MCP servers and clients
  39. 39Context engineering
  40. 40Human approval steps
  41. 41Guardrails
  42. 42Permissions and sandboxing
  43. 43Loop limits, budgets and kill switches
  44. 44How agents fail
  45. 45Advanced harness engineering
  46. 46Harnesses for long-running agents
  47. 47Serving an AI app
  48. 48Kubernetes and Helm basics
  49. 49One cloud in depth
  50. 50CI/CD with GitHub Actions
  51. 51Environments, secrets and Terraform
  52. 52Deploy and monitor an LLM app
  53. 53Tracing LLM apps
  54. 54Logs and traces for agents
  55. 55Online evals and feedback
  56. 56LLM gateways
  57. 57Inside LLM inference: prefill, decode and the KV cache
  58. 58Running models on many GPUs: replicas, tensor and pipeline parallelism
  59. 59Grow an AI app in layers: context, guardrails, routing, caching and agents
  60. 60Caching and speed
  61. 61Cost control
  62. 62Batching and model tiering
  63. 63Latency budgets and load testing
  64. 64Single sign-on
  65. 65Permission-aware RAG
  66. 66Audit logs
  67. 67Enterprise connectors
  68. 68Data pipelines for AI
  69. 69Data quality and personal data
  70. 70Defend against prompt injection
  71. 71Security and privacy
  72. 72Compliance for AI systems
  73. 73Security questionnaires
  74. 74Deployment models
  75. 75Open-weight models with vLLM
  76. 76Cost of ownership
  77. 77Discovery
  78. 78The KPI that matters
  79. 79Writing a scope of work
  80. 80Proof of concept, pilot, production
  81. 81Change management and training
  82. 82Demos, risk logs and escalation
  83. 83Talking to executives
  84. 84Managing scope creep
  85. 85Documentation and handover
  86. 86A case study, end to end
  87. 87Design an AI system
  88. 88The design interview
  89. 89The FDE interview loop
  90. 90Portfolio and job search

On the roadmap: Forward Deployed Engineer (FDE)

Client deliveryPro 3: AI system designerVideo lessonsRisk logEscalationStage gateKPI

Track

Today's AI turns are used

Lessons, roadmaps, videos, practice cards and the code playground stay free with your account, with no limit.