Skip to content
POLARISAI ACADEMY
Route R08

Local AI and Own Models

The Local AI and Own Models route has four courses that build on each other, from the basics to a final project of your own.

The four courses

  1. R08-101

    Running Models Locally: Hardware, Quantization and Engines

    The student can calculate exact VRAM and KV-cache memory requirements for any model architecture and quantization level, and configure an offline local inference server that passes API contract and determinism tests.

    22 h · Prerequisites: F01–F07, R02-101 · first lesson open

  2. R08-201

    Dataset Curation, Cleaning and Decontamination

    The student can build a deterministic dataset curation pipeline that deduplicates rows, strips PII, validates ChatML formatting, and enforces zero n-gram contamination between train and test splits.

    24 h · Prerequisites: R08-101, R02-201 · first lesson open

  3. R08-301

    Fine-Tuning with LoRA and QLoRA

    The student can implement a reproducible LoRA training and evaluation script with fixed random seeds that reduces validation loss on a toy task while passing catastrophic-forgetting regression checks.

    26 h · Prerequisites: R08-201 · first lesson open

  4. R08-401

    Benchmark Integrity and Air-Gapped Local Deployment

    The student can detect benchmark contamination in a supplied evaluation set and package a complete air-gapped local inference and retrieval service that passes all functional tests under strict firewall drop rules.

    26 h · Prerequisites: R08-301, R03-201 · first lesson open

What has to be passed

Every course ends with a task marked against a fixed answer key.

  • R08-101 Pass a 25-item keyed VRAM, bandwidth, and quantization calculation benchmark, and submit a local inference configuration and benchmark script that passes 20 automated tests in a zero-network (--network=none) container with a tiny fixture model.
  • R08-201 Submit a data curation pipeline (curate.py) tested against a seeded 5,000-row messy JSONL fixture containing duplicates, broken schemas, PII, and leaked test items; automated tests verify 100% clean output splits and zero train/test leakage.
  • R08-301 Submit a PyTorch/PEFT LoRA training and evaluation script executed on a tiny CPU/GPU fixture model in the isolated runner; automated tests verify trainable parameter count matches exact LoRA math (r * d), loss decreases deterministically, and adapter export succeeds.
  • R08-401 Complete a keyed contamination audit of a seeded benchmark dataset and submit a containerised local AI + retrieval stack that passes 30 end-to-end functional tests with outbound networking completely disabled.
Lessons coming

The lessons for this course are being written. Join the waitlist and we will email you when it opens. Join the waitlist