هوش مصنوعی برای تحلیل‌گران کسب‌وکار: مدیریت نیازمندی‌ها، مدل‌سازی و تحلیل داده

عنوان دوره : Udemy – AI for Business Analysts: Requirements, Models and Data

سطح : متوسط

زمان کل دوره : 1 ساعت و 38 دقیقه

تعداد فايل ها : 35 (1080p)

سال آموزش : 2026

فایل تمرینی : دارد: ZIP (1)

مربی دوره : Laurentiu Raducu

نرم افزار : –

زیرنویس انگلیسی : ندارد

Udemy – AI for Business Analysts: Requirements, Models and Data

Business analysis is mostly reading, listening and writing: interviews, workshop notes, requirements, process maps and status updates. AI is very good at a lot of that work, and dangerously bad at some of it. This course shows you which is which, on one realistic project, with real results.

You follow a freight company’s Delivery Exceptions Portal from the first brief to a prioritised backlog. You get the brief, three stakeholder interviews, a workshop transcript, a claims policy, a vendor offer, a draft requirements document and a quarter of ticket data. And the project has traps on purpose: two stakeholders who want opposite things, phrases that sound good and mean nothing, open tickets that distort a careless average, and a document with a hidden instruction for AI.

What you’ll do with AI:

Prompting that works: context, task, format and limits, shown on the same interview summarised twice.

Elicitation: interview guides, summaries with exact quotes (checked word for word), conflicts between stakeholders, questions for a vague brief, and minutes with real owners and dates.

Requirements: functional requirements with sources, user stories with Given/When/Then acceptance criteria, a strict ambiguity review, non-functional requirements and a traceability table.

Processes and models: as-is and to-be processes as Mermaid diagrams, gap analysis, decision tables from a policy, and a stakeholder map.

Data analysis: AI that runs code instead of guessing, SQL explained in plain English, and every number recomputed independently.

Documents: a requirements document section, one update for two audiences, MoSCoW prioritisation and a backlog ready to import.

Risks: testing for invented facts, redacting confidential data before prompting, hidden instructions in documents, and inconsistent answers.

Your way of working: a prompt library, a nine-point review checklist, and what stays your job.

Real prompts, real answers. Every prompt was sent to Claude and every answer is shown as it came back, with the time and the cost: 32 recorded runs, for less than one dollar in total. When a result surprised me, you see that too, such as the plausible additions that slip into otherwise excellent answers, and the data error Claude found that I hadn’t planted.

No programming needed. All case study files are included as a download, so you can rerun every prompt yourself.

Recorded with Claude Sonnet, October 2026. About 1 hour 40 minutes in 35 lectures of two to four minutes, each with a short recap.

What you’ll learn:
– Write prompts with context, task, format and limits that turn interviews, transcripts and documents into usable analysis
– Prepare interviews, summarise them with exact quotes, find conflicts between stakeholders and turn workshops into minutes
– Draft and review requirements, user stories with acceptance criteria, non-functional requirements and traceability tables
– Map as-is and to-be processes with Mermaid, write gap analyses and decision tables, and build a stakeholder map
– Analyse data with AI that runs code instead of guessing, and check every number and its interpretation yourself
– Spot and handle the risks: invented facts, hidden instructions in documents, confidential data and inconsistent answers
– Build a prompt library and a review checklist so you can use AI on real projects responsibly and consistently

Requirements:
– Some experience with business analysis, product ownership or requirements work is helpful, but not required
– No programming knowledge needed. A few lectures show Claude running code; you learn to read and check the result
– To try the prompts yourself: access to Claude or a similar AI assistant. Check your organisation’s policy before using real project data

Who this course is for:
– Business analysts who want to use AI for elicitation, requirements, modelling and documentation, without losing rigour
– Product owners and product managers who write user stories, acceptance criteria and backlogs
– Project managers, consultants and process analysts who turn interviews and workshops into decisions and documents
– Anyone who wants to see what AI really does with project documents: real prompts and answers, checked against the sources