Connecting to live activity…

LLM (Large Language Models) Bootcamp

Training concepts, fine-tuning, alignment, and inference for builders who go beyond API calls.

9 weeks·Hybrid·Cohort format

Overview

Understand how large language models work under the hood: pre-training objectives, fine-tuning strategies, alignment techniques, and efficient inference. This bootcamp balances theory with hands-on exercises so you can reason about cost, quality, and safety.

Who it's for

  • Engineers moving from application layer to model-aware roles
  • Researchers entering industry with shipping expectations
  • Senior ICs who need credible LLM depth in interviews

What you will learn

Curriculum breakdown, from foundations to portfolio-ready work.

Phase 1

Transformers & training

Mechanics that interviews probe.

  • Attention, scaling, and context windows
  • Pre-training objectives and data mixtures
  • Compute and memory trade-offs at a high level
Phase 2

Fine-tuning & alignment

When and how to specialize models.

  • Supervised fine-tuning and instruction tuning
  • RLHF and preference optimization (conceptual + practical)
  • Evaluation harnesses and red-team basics
Phase 3

Inference & systems

Fast, reliable LLM services.

  • Quantization and serving considerations
  • Batching, KV cache, and latency
  • Capstone: fine-tune or adapt a small model with eval story

Key features

  • Hybrid cohort with live Q&A
  • Guided readings + implementation labs
  • Mentor reviews on experiment write-ups
  • Mock interviews on LLM architecture and trade-offs
  • Assignments with reference solutions

Projects

Instruction tuning mini-project

Adapt a small open model with a clean dataset and eval set.

Inference benchmark

Compare latency/quality across settings and document findings.

Outcomes

Skills gained

Conceptual and practical depth across training, alignment, and serving.

Job readiness

Answer tough LLM questions with structured reasoning.

Portfolio

Technical write-ups that signal seniority.

Your mentor

Dr. Rohan Iyer

Research Engineer · LLM infrastructure

PhD + 8 years in NLP and large-scale training

Rohan bridges research intuition with production constraints: latency budgets, eval discipline, and honest limitations.

TransformersFine-tuningServing

Duration & schedule

Total duration

9 weeks

Weekly commitment

12–16 hours

Cohort format

Two live sessions weekly; heavy reading and labs async.

Ready to join the next cohort?

Secure your seat or request the full syllabus. We'll confirm prerequisites and start dates.

← All bootcamps