Lesson 12 of 12 intermediate

Career evidence and a defensible capstone

CV-তে term না, নিজের বোঝা কাজ আর প্রমাণ দেখাও। অন্য কোম্পানির নামে নিজের কাজ দিও না।

Guess first

Pick an answer. You will see if you were right at the end of the lesson.

Would the employer and the evidence confirm the claim?

Open interactive version (quiz + challenge)

Real-world analogy

A portfolio is a demonstration of how you solve problems, not a shelf of impressive tool names.

What is it?

A strong case study connects a user problem, constraints, decisions, failures and verified results. It distinguishes personal work from employer work and learning from production delivery. No course or ATS format guarantees an interview.

Real-world relevance

One shared support-ticket product, three optional entry points. Frontend: Show an accessible feature, a reproduced browser bug and the evidence behind its fix. Backend: Show an authorised API, a failed request case and a tested recovery decision. Mobile: Show offline recovery, a device-specific failure and the evidence behind the fix.

Key points

Choose a clear role — Keep your strongest professional lane visible, whether frontend, backend or mobile. Add AI-assisted work with evidence rather than renaming yourself an expert in every stack.
Keep attribution truthful — Label independent products and employer work separately.
Use measured outcomes — Include sample size, environment and limitations when quoting improvements.
Explain tradeoffs — Show why one task uses a rule and another uses a model.
Show a failure — A reproduced failure and tested fix demonstrate judgement.
Protect confidentiality — Use synthetic examples and only public or permission-cleared employer facts.
Practice without assistance — Explain the architecture, debug a failing case and answer cost/security questions without generated prompts.
Frontend practice — Show an accessible feature, a reproduced browser bug and the evidence behind its fix.
Backend practice — Show an authorised API, a failed request case and a tested recovery decision.
Mobile practice — Show offline recovery, a device-specific failure and the evidence behind the fix.

Quick check

Answer from memory — scrolling up first makes it stick less.

What artifact demonstrates this lesson?

Which statement is defensible?

Code example

Case study
Problem: duplicate offline support tickets
My role: independent learning project
Decision: persist idempotency key; review AI drafts
Verification: attach actual test report
Cost comparison: attach measured runs, or say unavailable
Limitation: fixture mode is not production inference
Next step: authorised sandbox evaluation

Line-by-line walkthrough

1. This template prevents a learning project from becoming an invented employer achievement. Replace placeholders with artifacts, not estimates presented as results.

Spot the bug

Built enterprise AI orchestration at my employer (actually a personal tutorial)

Need a hint?

Would the employer and the evidence confirm the claim?

Show answer

Attribute the work to an independent project and describe exactly what you built and verified.

Explain like I'm 5

Fun fact

A useful artifact for this lesson: Keep your strongest professional lane visible; extend it with evidence.

Hands-on challenge

Choose ONE stack; you do not need to build all three. Junior foundation: Explain one task, your contribution and a verified result without AI help. Senior extension: Defend an architecture decision, an incident lesson and an evidence-based cost tradeoff. Evidence: save the input, expected behaviour, actual result and one failure or limitation. Junior means basic coding and Git, not zero programming experience.

More resources

Building effective agents (Anthropic)
Download the free offline lab (Python, tests and instructions) (Teamz Lab)

Open interactive version (quiz + challenge) Download the whole course as PDF / Word

Lock it in

← Back to course: AI-Assisted Software Engineering