LLMOps & Inference at Scale

Serving frameworks, batching, caching, quantisation, and the cost / latency trade-offs that decide whether an LLM project is viable.

⏱ 6 weeks🎚 Advanced🧩 2 projects · Certificate🖥 Any device · our GPUs

Apply for this course Where this leads

Draft syllabus. Timeline and topics are indicative and being finalised with the teaching team — the shape is right, the week-by-week detail may shift.

🎯 Who it's for

Platform or MLOps engineers running LLM workloads in production.

✅ Prerequisites

MLOps on the Cloud or equivalent
Comfort with GPUs and containers
Basic understanding of transformer inference

Topics covered

Serving frameworks and continuous batching
KV cache and memory management
Quantisation and distillation

Caching and request routing
Autoscaling GPU fleets
Cost / latency / quality trade-offs

Expected completion timeline

6 weeks part-time at 8–12 hours per week ≈ 1.4 months. Self-paced learners can go faster; the live cohort keeps this pace.

Wk 1–2

Serve efficiently

You produce: An LLM endpoint with batching and caching, benchmarked.

Wk 3–4

Shrink and scale

You produce: A quantised deployment with an autoscaling policy.

Wk 5–6

Cost model and project

You produce: A cost / latency report and a production-ready config.

🧭 Where this fits

Advances the MLOps Engineer path toward senior LLMOps roles. See the career & salary map for the full path, the expected pay by region, and the total cost.

💳 Delivery & fees

Available in all four delivery formats. Fees vary by format and are confirmed on your advisor call; instalments available. [Add the fee and the next cohort date here.]

Add this to your plan

Book a call and we'll place this course in a full path toward the role you want.

Talk to an advisor