1
Curious builder0 XP earned · 300 to level 2
0 daysFinish a lesson to begin
Badge collection0 of 6 unlocked
51 small wins to finish your path
Behavioural: "Tell me about a project that failed or didn't go as planned." How do you answer well?
30-second answerSay your answer out loud first, then reveal.
Strong example answer (illustrative)
Situation: "I led a 4-week POC for an AI document assistant at a logistics company. We built a strong demo on sample PDFs."
What went wrong: "In week 3 we got access to their real document store and discovered 40% were scanned images in mixed languages. Our accuracy dropped from 88% to 61%, and we missed the go/no-go date. The root cause was mine: I accepted sample data at kickoff and didn't push for a real data inventory in week 1."
Response: "I told the sponsor immediately with a breakdown by document type, proposed a two-week extension focused on OCR and a scope limited to typed documents, and we reached 86% on that subset. The customer proceeded with a smaller pilot, but we lost the larger expansion that quarter."
Lessons applied: "Now every engagement I run starts with a data inventory on real production samples within the first three days, and I put data access as a formal exit criterion for kickoff. On my next two engagements, that surfaced data issues on day 2 instead of week 3."
What went wrong: "In week 3 we got access to their real document store and discovered 40% were scanned images in mixed languages. Our accuracy dropped from 88% to 61%, and we missed the go/no-go date. The root cause was mine: I accepted sample data at kickoff and didn't push for a real data inventory in week 1."
Response: "I told the sponsor immediately with a breakdown by document type, proposed a two-week extension focused on OCR and a scope limited to typed documents, and we reached 86% on that subset. The customer proceeded with a smaller pilot, but we lost the larger expansion that quarter."
Lessons applied: "Now every engagement I run starts with a data inventory on real production samples within the first three days, and I put data access as a formal exit criterion for kickoff. On my next two engagements, that surfaced data issues on day 2 instead of week 3."
What makes it strong
- Ownership of a specific mistake.
- Quick, transparent communication.
- A partial recovery with numbers.
- A durable process change, with evidence it worked later.
Related
Every expert started right here.