Case Study: Shipping a Tool-Using LLM Assistant with Built-In Verification Under Latency Constraints
You are the product owner for an internal LLM assistant used by Customer Operations to answer: (1) “Where is order #12345 right now?” and (2) “Can I promise delivery by Friday?” The assistant can call two external APIs: get_tracking(order_id) (returns latest scan location + timestamp) and get_inventory(sku, warehouse) (returns available-to-promise quantity). A recent incident occurred: the assistant confidently promised Friday delivery based on a stale tracking scan and an incorrect assumption about inventory allocation rules. Leadership now requires: (a) fewer than 2 API calls per user request on average, (b) a measurable reduction in incorrect commitments, and (c) an auditable record of why the assistant made a commitment.
As the designer, propose a single end-to-end inference workflow that integrates: deliberate analysis before answering, self-reflection, predict-then-verify with a verifier, and tool use with the APIs above. Your answer must specify (i) when and why the model calls each API (or chooses not to), (ii) what the model generates as multiple “candidates” (what varies across candidates), (iii) what the verifier checks and what evidence it uses (including how it handles stale timestamps), and (iv) what the final user-facing response should contain to be auditable while minimizing overconfident promises.
0
1
Tags
Ch.3 Prompting - Foundations of Large Language Models
Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Ch.5 Inference - Foundations of Large Language Models
Related
Comparison of Execution Timing in Tool Use and RAG
Limitation of Pre-trained LLMs in Tool Use
Using Web Search as an External Tool for LLMs
Application of LLMs to Mathematical Problems
A development team is building a chatbot for an airline. The chatbot must be able to answer user questions like, 'What is the status of flight UA456?' and 'Are there any business class seats available on the 10 AM flight to London tomorrow?'. The airline's flight data is stored in a private, real-time database. Which of the following represents the most effective and reliable approach for the team to implement these features?
A user asks a Large Language Model, 'What is the capital of Brazil and what is the current time there?'. The model has access to an external tool
get_current_time(city). Arrange the following steps in the logical order the model would follow to answer the user's request.Troubleshooting an LLM-Powered Sales Assistant
Designing a Reliable LLM Workflow for Real-Time Decisions
Design Review: Combining Tool Use, DTG, and Predict-then-Verify for a High-Stakes API Workflow
Post-Incident Analysis: Preventing Confidently Wrong API-Backed Answers
Case Review: Preventing Incorrect Refund Commitments in an LLM + Payments API Assistant
Case Study: Shipping a Tool-Using LLM Assistant with Built-In Verification Under Latency Constraints
Case Study: Preventing Hallucinated Compliance Claims in an API-Enabled LLM for Vendor Risk Reviews
You are reviewing a proposed architecture for an i...
In an LLM-based customer support assistant, the mo...
You’re designing an internal LLM assistant for a f...
You’re leading an internal rollout of an LLM assis...
Methods for Activating Self-Reflection in LLMs
An AI model is asked, 'What is the approximate distance from the Earth to the Moon?' It provides two consecutive responses:
- Response 1: 'The distance from the Earth to the Moon is about 238,900 kilometers.'
- Response 2: 'Upon review, my previous answer was imprecise. The distance is in miles, not kilometers. The correct average distance is approximately 238,900 miles, which is about 384,400 kilometers. Stating the unit correctly is crucial for accuracy.'
Which of the following best analyzes the process demonstrated in Response 2?
Evaluating AI Response Quality
Mechanism of AI Self-Correction
You are reviewing a proposed architecture for an i...
You’re designing an internal LLM assistant for a f...
You’re leading an internal rollout of an LLM assis...
In an LLM-based customer support assistant, the mo...
Design Review: Combining Tool Use, DTG, and Predict-then-Verify for a High-Stakes API Workflow
Designing a Reliable LLM Workflow for Real-Time Decisions
Post-Incident Analysis: Preventing Confidently Wrong API-Backed Answers
Case Study: Shipping a Tool-Using LLM Assistant with Built-In Verification Under Latency Constraints
Case Review: Preventing Incorrect Refund Commitments in an LLM + Payments API Assistant
Case Study: Preventing Hallucinated Compliance Claims in an API-Enabled LLM for Vendor Risk Reviews
Limitation of the Deliberate-then-Generate (DTG) Method
Comparison of Iterative vs. Non-Iterative Prompting Methods
Instructional Component of the DTG Prompt Template for Translation Refinement
Integration of Feedback and Refinement in the DTG Method
A developer is using a Large Language Model to refine a technical summary. They want the model to first identify any factual inaccuracies or unclear statements in the original text and then, based on that analysis, produce a corrected and more coherent version. Which of the following approaches correctly implements the 'Deliberate-then-Generate' method for this task?
Input Structure of the DTG Prompt for Chinese-to-English Translation
Challenge of LLM-Based Error Identification in Translation
A developer is designing a workflow to refine user-generated reports using a Large Language Model. The primary goal is to ensure the model first analyzes potential issues (e.g., ambiguity, factual errors) before rewriting the report, all while minimizing the number of interactions with the model. Which of the following prompt structures best represents the 'Deliberate-then-Generate' method for this task?
Analysis of a Translation Refinement Process
You are reviewing a proposed architecture for an i...
You’re designing an internal LLM assistant for a f...
You’re leading an internal rollout of an LLM assis...
In an LLM-based customer support assistant, the mo...
Design Review: Combining Tool Use, DTG, and Predict-then-Verify for a High-Stakes API Workflow
Designing a Reliable LLM Workflow for Real-Time Decisions
Post-Incident Analysis: Preventing Confidently Wrong API-Backed Answers
Case Study: Shipping a Tool-Using LLM Assistant with Built-In Verification Under Latency Constraints
Case Review: Preventing Incorrect Refund Commitments in an LLM + Payments API Assistant
Case Study: Preventing Hallucinated Compliance Claims in an API-Enabled LLM for Vendor Risk Reviews
Supervised Learning of Verifiers
Relation between Verifiers and RLHF Reward Models
Classification of Verification Approaches
Guiding Role of the Verifier in Self-Refinement
A system is designed to solve complex, multi-step logic puzzles. First, a generative model produces five different potential step-by-step solutions to a given puzzle. Then, a second, distinct component is used. This second component's sole function is to evaluate each of the five proposed solutions by scoring the logical soundness of each step in the reasoning chain. Based on these scores, it selects the single most coherent and valid solution to present as the final answer. What is the primary role of this second component in the system's architecture?
Improving an AI Tutoring System
Consider a system that solves a problem by first having one component generate several different step-by-step solutions. For this system to be effective, the same component that generated the solutions must also be used to evaluate them and select the best one.
You are reviewing a proposed architecture for an i...
You’re designing an internal LLM assistant for a f...
You’re leading an internal rollout of an LLM assis...
In an LLM-based customer support assistant, the mo...
Design Review: Combining Tool Use, DTG, and Predict-then-Verify for a High-Stakes API Workflow
Designing a Reliable LLM Workflow for Real-Time Decisions
Post-Incident Analysis: Preventing Confidently Wrong API-Backed Answers
Case Study: Shipping a Tool-Using LLM Assistant with Built-In Verification Under Latency Constraints
Case Review: Preventing Incorrect Refund Commitments in an LLM + Payments API Assistant
Case Study: Preventing Hallucinated Compliance Claims in an API-Enabled LLM for Vendor Risk Reviews
Verifiers in LLM Reasoning
The Predict-then-Refine Paradigm in NLP
Self-Refinement in LLMs
Generating and Verifying Thinking Paths
Solution Selection as a Search Problem
Reasoning Path in Problem Solving
Best-of-N Sampling (Parallel Scaling)
Comparison of Parallel Scaling and Self-Refinement
Verifier
Solution as a Sequence of Reasoning Steps
A team is developing a system to solve complex mathematical word problems using a large language model. Their goal is to maximize the final answer's accuracy. Which of the following strategies best exemplifies a process where multiple potential solutions are first generated and then evaluated to select the most reliable one?
Analyzing LLM Reasoning Strategies
A system is designed to solve a complex problem by first generating multiple possible answers and then selecting the best one. Arrange the following steps to accurately represent this two-stage workflow.
In a system designed to solve a problem by first generating multiple potential solutions and then using a separate component to select the best one, the quality of the final selected answer depends solely on the generative capability of the initial model.
You are reviewing a proposed architecture for an i...
You’re designing an internal LLM assistant for a f...
You’re leading an internal rollout of an LLM assis...
In an LLM-based customer support assistant, the mo...
Design Review: Combining Tool Use, DTG, and Predict-then-Verify for a High-Stakes API Workflow
Designing a Reliable LLM Workflow for Real-Time Decisions
Post-Incident Analysis: Preventing Confidently Wrong API-Backed Answers
Case Study: Shipping a Tool-Using LLM Assistant with Built-In Verification Under Latency Constraints
Case Review: Preventing Incorrect Refund Commitments in an LLM + Payments API Assistant
Case Study: Preventing Hallucinated Compliance Claims in an API-Enabled LLM for Vendor Risk Reviews
Sequential Scaling