Decagon Applied AI Engineer interview questions

Token count is latency and dollars, not just style. 'Be concise' barely works, and max_tokens just truncates mid-sentence. Here is how to actually shape length without amputating answers.

Agents are over-applied. The strong answer resists the hype: most tasks want a single structured call, and agents earn their cost only under specific conditions. Here is when to reach for one and how to stop it from compounding errors.

Conversations and agent tasks outgrow the context window, and naive 'stuff the whole history' fails on cost, latency, and lost-in-the-middle. The signal is a tiered memory design: recent buffer, summarized mid-term, retrieved long-term.

Agents need memory beyond the context window, and 'just store the history' is the wrong answer. The signal is distinguishing the memory types and mapping each to a storage and retrieval strategy. Here is the answer.

An agent that retries the same failed action 40 times is a runaway bill and a stuck user. Termination is something you engineer into the harness, not something the model reliably decides. Here is the control layer.

Real tools time out, rate-limit, and return malformed JSON. An agent that assumes every call succeeds is a demo, not a product. Here is the error-handling layer that keeps it alive in production.

Real users type 'what about the new policy?' with no context. Retrieving on those four words returns noise. Here is how strong systems disambiguate before they ever hit the retriever.

When a chain of LLM calls and tools produces a wrong final answer, 'the model was bad' is not a diagnosis. You need to see every step. Here is the tracing layer that turns a black box into something debuggable.

Frameworks promise speed and deliver abstraction you'll eventually fight. Rolling your own is more code but full control. Here is the honest tradeoff, and why many production teams end up thinner than they started.

The industry conflates 'agent' with anything that calls an LLM. The useful distinction is how much control you hand to the model, and more autonomy is not better. Here is the spectrum and the decision rule.

A live-coding round that separates people who have shipped agents from people who have read about them. The signal is a real plan-act-observe loop, tools that hit an actual SQL database, guardrails that stop the obvious failures, and a concrete evaluation plan, not a single prompt that pretends to be an agent.

A context window is not memory. Real agents need a storage architecture: working buffer, episodic recall, and consolidated facts. The signal is the read/write paths and how you keep memory from rotting.

Autonomous agents still need humans at the right moments. The signal is designing checkpoints: where to interrupt, what to surface, how to persist and resume state, without turning the agent into a click-through rubber stamp.

Single-turn evals cannot catch failures that history causes, because they have no history. The offline answer is two suites, one deterministic and one simulated, and an honest account of why the simulated user will lie to you about how good your agent is.

Customer support is the killer applied-AI use case, testing agents, RAG, tools, and the all-important human escalation. The signal is knowing when to resolve, when to act, and when to hand off, safely.

An AI support agent handling millions of chats will sometimes be wrong, rude, or unhelpful, and you cannot read them all. Monitoring quality at that scale is its own design problem with its own failure modes.

Auto-resolving support tickets means grounding answers in your knowledge base, safely calling real APIs (refunds, account changes), and knowing when to hand off to a human. Learn the agent loop, the guardrails, and how to measure resolution without breaking trust.

Hype sets customers up to expect a system that reads minds and never errs. Resetting that without losing the deal is an applied-AI skill interviewers probe directly. Here is the move.

AI features fail in public, sometimes embarrassingly. How you respond to the customer, not just the bug, is what this question is really testing. Here is the recovery that rebuilds trust.

An angry customer is a test of composure and ownership, not just empathy. Interviewers want to see you de-escalate, fix the real problem, and rebuild trust. Here is the arc that scores.

Saying no to a customer is a relationship skill: protect them from a bad outcome while keeping their trust. Interviewers want the reason, the alternative, and the framing. Here is the move.

Almost everyone can quote softmax(QKᵀ/√d_k)V. The interviewer wants the variance argument and the exact training failure the scale prevents. Here is the answer that separates memorization from understanding.

The single most-asked applied GenAI question, and the one most candidates answer as a definition dump. The interviewer wants a decision framework with a default and the conditions that flip it. Here is the call that scores.

The modal Applied AI design round. Anyone can draw embed-retrieve-generate. The signal is in chunking, hybrid retrieval, the rerank/latency tradeoff, and how you prove it works. Here is the structure that scores.

MoE is why some frontier models have huge parameter counts but serve cheaply. The signal is the sparse-activation idea, the routing mechanism, and the operational cost (memory, load balancing) that the FLOP savings quietly hide.

These are the knobs every LLM product tunes, and the interviewer wants to know what each does to the probability distribution and which setting fits which task. Including why factual and creative tasks want opposite settings.

The two knobs that make or break a RAG system, and the ones candidates hand-wave. The signal is tuning chunking against recall and knowing exactly what dense retrieval misses that BM25 catches. Here is the reasoning, not the rules of thumb.

Applied AI rarely waits for certainty, and companies (Amazon's Bias for Action, NVIDIA's speed) test whether you can act decisively without freezing or being reckless. The signal is a reversible, well-reasoned call with a plan to validate. Here is the arc that lands.

Alignment is now table stakes even outside 'safety' roles. The interviewer wants the three-stage RLHF pipeline and a crisp account of why DPO dropped the reward model entirely. Here is the answer with the mechanism, not the buzzwords.