108. Design an LLM Inference Serving Platform

The brief

Design an LLM Inference Serving Platform. Open-weight LLMs from 8B to 405B served to many products over a streaming OpenAI-style API on AWS GPUs. Work from the scoping questions below. State assumptions for any unspecified load, guarantee or target, then trace your design end to end. Explain one difficult case and a credible alternative; the worked example is a reference, not a required implementation.

• Set the scope: Traffic? Targets?
• Define the contract: Conversations? Tenants?
• Test the boundaries: Batch? Context length? Smaller models?

Constraints

Explicit scope and guarantees

Resolve the scoping questions for an LLM Inference Serving Platform. Separate stated behavior from assumptions, and identify what is outside your design.

Supported operating targets

Declare relevant volume, latency, freshness, quality or cost targets with units. Show calculations or an evaluation plan that can test them; unspecified targets are your assumptions, not hidden pass criteria.

Failure and boundary behavior

Explain how your guarantees hold in a difficult case relevant to this subject. Address: Context length? Smaller models?

What to cover

1. 01

   Scope and behavior contract

   Identify users, required behavior and exclusions. Answer: Traffic? Targets?

2. 02

   State and interfaces

   Define the information owned by the system and the inputs, outputs and errors at its boundaries. Resolve: Conversations? Tenants?

3. 03

   Capacity and operating targets

   Estimate the dominant workload and resource demand with units and explicit assumptions. For a learned system, also state how quality is measured and what data is available.

4. 04

   Architecture and central flow

   Draw or describe the responsibilities needed for an LLM Inference Serving Platform. Trace a representative request, event or job from its input to a visible result; identify durable state owners.

5. 05

   Failure and boundary walkthrough

   Walk through a difficult case step by step, including detection and recovery. Consider: Batch? Context length? Smaller models?

6. 06

   Tradeoffs and operations

   Compare a credible alternative using your chosen workload and guarantees. Explain a remaining risk, a signal to watch and when you would change the design.

Worked designs

Explore the architecture and decisions, then build on an example with Coach.

Official design

Design an LLM Inference Serving Platform

Open-weight LLMs from 8B to 405B served to many products over a streaming OpenAI-style API on AWS GPUs: one-hour boards for junior, senior and staff, with the theory behind them.

Explore the design

Review rubric

AI feedback uses these criteria. Scores are practice feedback.