REALLIER AI LEARN · 5 SPECIALIZED TRACKS · SUB-TRACK DIRECT ACCESS · PRACTICAL OUTPUTS

Explore AI,starting from a real task today

Organized into five core tracks: AI Office & Productivity (16 lessons), AI Principles & Labs (16 code sandboxes), AI 3D Modeling & Assets (16 lessons), AIGC Visual & Media (ControlNet, ComfyUI, Video/TTS, AI Comic Drama), and AI Finance & Research.

Five Core Learning Tracks (with 20 Sub-Track Shortcuts)

3 tracks are live with 48 interactive lessons; AIGC Visual & Media (including AI Comic & ComfyUI) and AI Finance tracks are under construction
Live · 16 Lessons
Everyday Workplace & Data Delivery

AI Office & Productivity

Start from real tasks: master briefing, verification, email, meeting notes, table cleaning, chart reading, and reports.

Explore Track →
Live · 16 Code Labs
Interactive Model & Agent Code Sandboxes

AI Principles & Labs

Zero-video live code sandboxes: neural function fitting, Transformer ablation, MoE & 1M KV-cache calculators, and raw Agent/MCP/Skill payload inspectors.

Explore Track →
Coming Soon · 16 Planned
ControlNet · ComfyUI · Video/TTS · Comic Drama

AIGC Visual & Media

Unifies 2D controlled generation, ComfyUI commercial visuals, AI video/voice synthesis, and AI comic drama into four specialized sub-tracks.

Preview Sub-Tracks →
Live · 16 Lessons
3D Gen · Retopology · PBR & GLB QA

AI 3D Modeling & Assets

From generated drafts to production-ready 3D assets: multi-provider generation, Blender bmesh topology QA, UV/PBR baking, and CAD-to-3D delivery.

Explore Track →
Coming Soon · 12 Planned
Filings Audit · Time-Series · Quant Assist

AI Finance & Research

For financial research and quant workflows: filing citation extraction, bull/bear report cross-checking, time-series cleaning, and risk controls.

Preview Sub-Tracks →

AI Principles & Labs

Zero-video live code sandboxes: neural function fitting, Transformer ablation, MoE & 1M KV-cache calculators, and raw Agent/MCP/Skill payload inspectors.
01Principles & engineering

Everything is a function: from linear fit to giant classification

Tune weights and activations to fit curves, and switch output heads to see next-token prediction as 50,257-way classification

Open lesson
02Principles & engineering

How text becomes numbers: tokenization and high-dimensional embeddings

Compare tokenizer boundaries and drag the “Apple” vector in 2D space to compute live cosine similarity

Open lesson
03Principles & engineering

Training neural networks: loss, gradient descent, and parameter accounting

Tune learning rate to run live gradient descent and calculate exact Transformer layer parameter shares

Open lesson
04Principles & engineering

Attention mechanism: why a single word cannot predict the future

Compare 4 evolutionary stages of attention and drag the query vector to watch live weight pooling

Open lesson
05Principles & engineering

FFN and Linear Projection: the true compute and parameter workhorse

Toggle FFN expansion ratios (768→3072→768) and GELU activation to inspect feature filtering and vocab projection

Open lesson
06Principles & engineering

Positional encoding, residual, and norm: break the model via ablation

Flip off Positional Encoding, Residual, and Norm switches in the ablation lab to witness word-order blindness and deep signal collapse

Open lesson
07Principles & engineering

From Logits to the next token: Softmax, Temperature, and Top-P sampling

Adjust Temperature, Top-K, and Top-P thresholds and draw 20 live samples to see probability reshaping and tail pruning

Open lesson
08Principles & engineering

Pretraining & Few-Shot emergence: how next-token prediction builds a generalist

Generate parallel self-supervised training pairs from any sentence and switch 0/1/3-Shot prompts to observe in-context learning

Open lesson
09Principles & engineering

Post-training trio: SFT, Chain-of-Thought (CoT), and RLVR

Switch across Base, SFT, CoT, and RLVR stages on a business math problem to inspect output streams and accuracy

Open lesson
10Principles & engineering

Modern architecture evolution (Part 1): Sparse MoE and dynamic routing

Switch token domains in the MoE router sandbox to watch gated experts light up and audit compute savings

Open lesson
11Principles & engineering

Modern architecture evolution (Part 2): 1M context, KV Cache, and MLA

Drag the context slider (4K → 1M) and switch MHA, GQA, MLA, and Linear Attention to audit live KV Cache VRAM bills

Open lesson
12Principles & engineering

Compute, hardware, and quantization: GPU, SSD, and FP8/UE8M0

Switch precision formats and memory tiers (SSD, DDR5, HBM3e) in the sandbox to calculate live throughput and quantization error

Open lesson
13Principles & engineering

Demystifying chat & memory: stateless models, Context, and compression

Compare Stateless, Full-History Concat, and Summary Memory in the payload inspector to view the actual raw JSON sent to the model

Open lesson
14Principles & engineering

External knowledge & action protocols: RAG, Function Calling, and MCP

Toggle RAG evidence passages above and inspect Function Calling JSON interception and MCP discovery below

Open lesson
15Principles & engineering

Workflows, Skills, and Sub-agents: context isolation engineering

Switch across 4 orchestration patterns in the sandbox to compare main-window token consumption (48.5K vs 2.8K) and flexibility

Open lesson
16Principles & engineering

Frontier myth-buster: multimodal fingers, Computer Use, watermarks, and benchmarks

Experiment with vision patch compression, GUI accessibility tags, green-list watermark bias, and benchmark contamination in the 4-in-1 lab

Open lesson

Practice progress stays in this browser. Copy or download your work; save draft edits before leaving.

Practical Prompt & Task Templates

Each template includes complete sample materials and verification rules

Four-Element Task Brief Template

Lock audience, key facts, format, and unknown rules upfront.

#Briefing#Grounded

Internal Event & Notice Email Template

Subject + core schedule + deadline action + open placeholders.

#Email#PreSendCheck

Multi-Source Evidence Summary Template

Tag every claim with [Source A/B/C] and separate caps from spend.

#Reading#Citations

Meeting Notes to Action Table Template

Split into settled facts, named owners/deadlines, and open gaps.

#Meeting#ActionTable

Spreadsheet Cleaning & Audit Log Template

Deduplicate with a log, standardize ISO dates, and preserve blanks.

#Spreadsheets#AuditLog

16:9 Visual Card Brief Template

Enforce 3-tier hierarchy and forbid unconfirmed promotional badges.

#VisualBrief#Layout

Complete AI Knowledge Map (8 Capability Areas · 32 Topics)

From foundational concepts and prompting to deliverables, model principles, and governance
8 areas / 32 topics
A

AI literacy & judgment

Understand capabilities, choose tasks, and assess results.

3 available
  • A01AI and its products
  • A02Tasks and delegation
  • A03Checking reliability
  • A04Data, permissions, and tools
B

Working with AI

Give clear instructions and improve the result.

4 available
C

Everyday office work

From email and meetings to tables and presentations.

5 available
D

Research & knowledge

Find evidence, make comparisons, and retain knowledge.

2 available
  • D01Search and research questions
  • D02Reading, summaries, and evidence
  • D03Comparing options
  • D04Organizing knowledge
E

Content & multimedia

Turn ideas into writing, images, audio, and video.

5 available
F

Role-based projects

Combine skills in a complete workplace project.

1 available
  • F01Sales and support
  • F02Marketing and operations
  • F03Administration, HR, and training
  • F04Product, projects, and management
G

Workflows & small apps

Turn a reliable method into a repeatable workflow.

5 available
H

Principles & engineering

Explore model mechanisms and application development.

23 available
  • H01How models learn and work
  • H02Application foundations
  • H03Retrieval, tools, and agents
  • H04Evaluation, deployment, and maintenance