Forward Deployed Engineer interview question · Scoping an ambiguous customer problem · mid level
A customer says: "We want AI for our support team." What do you do in the first meeting?
Free to read in full. This is 1 of the 100 questions in the Forward Deployed Engineer interview pack, with 127 diagrams. Get the pack, ₹999 · $20
Why it is asked
This is the classic FDE scoping question. The interviewer checks that you ask before you build and that you end with a concrete next step.
The answer
In the first meeting I turn the vague ask into a specific problem, a number we want to move and a small first step, and I do not propose a solution yet.
I ask questions in six areas:
- Users: who exactly? For example 120 tier-1 agents (the first line who answer tickets), or customers helping themselves?
- Pain: where is time or money lost? Long handle time, a backlog, slow first response, too many escalations, or wrong answers?
- Workflow and tools: which help desk (Zendesk, Salesforce Service Cloud), and where knowledge lives (the knowledge base, or KB, in Confluence or SharePoint, and old tickets).
- Data: how many past tickets, which languages, how much personal data in them.
- Baseline: what are today's numbers, for example average handle time of 9 minutes and a customer satisfaction score (CSAT) of 4.1 out of 5?
- Constraints: security review, data location rules, budget, deadline and who signs.
Then I suggest two or three candidate use cases and rank them by value and risk: agent assist (the tool drafts a reply and a human agent approves every draft), ticket routing, or a customer-facing chatbot. I usually recommend starting with agent assist, because a human checks every answer, so a wrong draft costs a few seconds instead of a wrong promise to a customer.
I end the meeting with a one-page summary, a named owner on their side, a data access request and a date for the next session.
What the interviewer listens for
- Questions before solutions.
- Users, pain, workflow, data, baseline, constraints.
- Several candidate use cases ranked by value and risk.
- Prefers a human-in-the-loop start.
- Leaves with an owner, a data request and a date.
Follow-ups the interviewer may ask
- The CEO wants a customer-facing chatbot live in a month. Hint: show the risk, offer agent assist first as the path to a chatbot, with the eval data it creates.
- They cannot give you any baseline numbers. Hint: measure it yourself from a ticket export in week one.
Common mistakes
- Demoing a chatbot in the first meeting.
- Leaving without a measurable goal or a next step.
Written by Roni Das, who builds distributed systems and AI systems for a living.
One question is not enough for the interview.
This is 1 of the 100 questions in the Forward Deployed Engineer pack
Each question has why it is asked, a full answer, what the interviewer listens for, the follow-ups, the common mistakes and a diagram (127 diagrams in all). A PDF you keep, junior to staff level.
India: UPI, cards or netbanking through Razorpay. Elsewhere: PayPal. Download at once from My packs.
More free questions from this pack
- The customer's admin offers you a global admin account "to save time". What do you ask for instead, and why? (junior)
- Your connector writes tickets into the customer's system. A request timed out, the connector retried, and now there are duplicate tickets. How do you make the writes safe to retry? (mid)
- The customer wants RAG over 10 million documents, all behind their firewall. Size it for me on the whiteboard. (senior)
- One deployment of your AI platform must serve 50 business units of the same enterprise. How do you design multi-tenancy? (staff)
Free questions from the other packs
- What do temperature and top-p control, and what values would you use for data extraction versus brainstorming? (AI Engineering)
- What is hybrid search, and how do you merge keyword and vector results with Reciprocal Rank Fusion? (AI Engineering)
- Users ask "list all X" questions and RAG returns only five items. Why, and what do you do? (AI Engineering)
- How do you check that an LLM judge agrees with human reviewers? (AI Engineering)
- Cut the cost of an LLM feature serving 10 million requests a day by 60% without hurting quality. Walk through your plan. (AI Engineering)
- Why are LLMs bad at counting letters or reversing a word? (GenAI and LLM)
- Write out scaled dot-product attention with the shape of every tensor. (GenAI and LLM)
- How does speculative sampling keep exactly the target model's distribution? (GenAI and LLM)
- What is GRPO, and how do verifiable rewards train reasoning models? (GenAI and LLM)
- Design a system that answers questions from 300-page contracts accurately. (GenAI and LLM)
- What happens, step by step, when an LLM server handles one request? (Inference Engineering)
- Derive the KV cache size formula and work it out for Llama 3 70B. (Inference Engineering)
- What is PagedAttention? (Inference Engineering)
- When does speculative decoding stop helping or make things worse? (Inference Engineering)
- Design a service to serve Llama 3 70B at 10 QPS with latency targets. (Inference Engineering)
- Compute the cosine similarity of a = (3, 4) and b = (4, 3) by hand. What does the result mean? (Maths for ML)
- A batch of 32 MNIST images goes into a dense layer with 128 units. What are the shapes, and what is the rule for when two matrices can be multiplied? (Maths for ML)
- A disease affects 1% of people. A test catches 90% of sick people (sensitivity) and wrongly flags 9% of healthy people (false positive rate). You test positive. What is the chance you are sick? (Maths for ML)
- Why do we not use mean squared error for classification with a sigmoid output? Compare the gradients numerically. (Maths for ML)
- Compute scaled dot-product attention by hand for two tokens with d_k = 2. (Maths for ML)