AI Product Teardowns

    For engineers and researchers who build AI systems

    How real AI products work under the hood. Architecture decisions, tech stacks, and key engineering insights.

    GitHub Copilot

    Agent Platforms
    Microsoft·Coding Agent Platform
    View product

    Tech Stack

    Multi-model LLMs
    Agent mode
    Cloud agent
    GitHub Actions sandboxes
    MCP

    How it works

    Copilot is no longer just inline autocomplete. The current architecture spans IDE chat and completions, local agent mode, and a cloud agent that researches a repository, drafts a plan, edits a branch, runs tests in an ephemeral GitHub Actions environment, and can open or update pull requests.

    Key Insight

    The moat has shifted from completion latency to workflow ownership: GitHub controls issues, branches, reviews, Actions, permissions, usage metrics, and enterprise policy around the agent.

    ChatGPT

    Office Productivity
    OpenAI·General AI Workspace
    View product

    Tech Stack

    GPT-5.6
    Reasoning models
    Web search
    Deep research
    Agent mode
    Images
    Codex

    How it works

    ChatGPT has become a front door for multiple model and tool paths: fast everyday chat, reasoning-heavy work, web search with citations, deep research reports, file/data analysis, image creation, scheduled tasks, connected apps, and Codex for software work.

    Key Insight

    The product advantage is orchestration. ChatGPT packages model choice, memory, files, browsing, app permissions, agent execution, and creative tools into one consumer workflow instead of exposing each capability as a separate developer primitive.

    Midjourney

    Media & Entertainment
    Midjourney·Image Generation
    View product

    Tech Stack

    Image models
    Video models
    Web editor
    Discord workflow
    Style/character references

    How it works

    Midjourney has expanded from Discord-first image generation into a web creative workflow with image editing, reference-driven generation, and video animation from generated or uploaded stills. The public product is still a closed creative model stack rather than an API-led platform.

    Key Insight

    Midjourney's differentiation remains taste and creator workflow. Its strongest users are not buying a generic diffusion endpoint; they are buying aesthetic defaults, iteration speed, references, and community-shaped discovery.

    Cursor

    Agent Platforms
    Anysphere·AI IDE
    View product

    Tech Stack

    Agent Composer
    Cursor Agent
    Cursor CLI
    Tab model
    Repo context
    Model routing

    How it works

    Cursor now presents itself as a coding agent across desktop, CLI, cloud, review, and autocomplete surfaces. The workflow ranges from fast Tab predictions to targeted edits, repository-aware agent plans, shell/tool use, and pull-request handoff.

    Key Insight

    Cursor's product wedge is the autonomy slider: developers can stay in low-friction autocomplete, ask for scoped edits, or let an agent plan and change larger parts of the codebase without leaving the editor workflow.

    Perplexity

    Office Productivity
    Perplexity AI·AI Search
    View product

    Tech Stack

    Answer engine
    Sonar API
    Search API
    Agent API
    Comet browser
    Citation UX

    How it works

    Perplexity has evolved from cited answer search into a search-and-agent stack: consumer answers, developer APIs for Sonar/search/agents, and Comet, a Chromium-based browser with an assistant that can summarize, research, email, shop, and act across web pages.

    Key Insight

    The core design bet is that search becomes an action surface. Citations build trust for answers, while Comet tests whether the browser itself can become the execution environment for everyday web tasks.

    HyperCLOVA X

    Sovereign AI
    Naver·Korean Foundation Model
    View product

    Tech Stack

    THINK
    SEED
    Omni models
    Korean tokenizer
    CLOVA Studio
    Naver services

    How it works

    HyperCLOVA X is now a Korean-first model family rather than one monolithic LLM. NAVER positions THINK for heavier reasoning, SEED variants for efficient deployment, and omni/vision/audio capabilities for multimodal Korean workflows through CLOVA Studio and NAVER service integrations.

    Key Insight

    The defensible layer is local context: Korean data, token efficiency, cultural nuance, domestic customer cases, and distribution through NAVER Cloud and NAVER-owned products matter more than frontier-model size alone.

    Claude Code

    Agent Platforms
    Anthropic·Coding Agent Harness
    View product

    Tech Stack

    Terminal harness
    CLAUDE.md memory
    Skills (SKILL.md)
    Hooks
    Subagents
    MCP
    Headless / CI

    How it works

    A terminal-first coding agent: a model wrapped in a harness of tools (bash, file read/edit, search), project memory in a CLAUDE.md checked into the repo, skills as on-demand instruction folders, hooks that intercept tool calls, and subagents that fan work out. The same harness ships as the Agent SDK and runs headless in CI.

    Key Insight

    The product is the harness, not the model. The layer that compounds — CLAUDE.md, skills, hooks — lives in git alongside your code, so the agent gets better per-repository over time. A chat window can't accumulate that.

    DeepSeek

    Sovereign AI
    DeepSeek·Frontier Model Economics
    View product

    Tech Stack

    Sparse MoE
    Multi-head Latent Attention
    FP8 training
    RL-first reasoning (R1)
    Distilled small models
    Open weights (MIT)

    How it works

    A cost-first frontier architecture built under GPU export constraints: a highly sparse mixture-of-experts that activates a small fraction of parameters per token, Multi-head Latent Attention that compresses the KV cache, and FP8 training. R1 then showed reasoning could be trained with large-scale RL and distilled down into small open models — all released under MIT.

    Key Insight

    DeepSeek treated cost as the product. The architecture exists to make frontier-class tokens radically cheaper, and the open weights plus aggressive API pricing reset the entire industry's economics — every closed lab now prices against it.

    Solar LLM

    Sovereign AI
    Upstage·Korean Frontier Model
    View product

    Tech Stack

    Depth up-scaling lineage
    Solar Pro line
    Document AI pairing
    OpenAI-compatible API
    Open releases
    MSIT sovereign consortium

    How it works

    Korea's startup route to a frontier model: Solar began with depth up-scaling — stacking layers of a proven smaller model, then continuing pretraining — and grew into the Solar Pro line, the first Korean family Artificial Analysis recognized as frontier-class. Upstage sells it as a stack with Document AI (parsing, OCR), and holds one of the five MSIT sovereign-AI mandates.

    Key Insight

    Solar proves you don't need hyperscale compute to stay relevant — architectural pragmatism plus a wedge enterprises actually pay for. The chat model gets the benchmarks; the document stack gets the contracts.

    Kling

    Media & Entertainment
    Kuaishou·Video Generation
    View product

    Tech Stack

    DiT video model
    Image-to-video
    Lip sync & effects
    Creator tooling
    Kuaishou distribution
    API

    How it works

    A diffusion-transformer video model shipped inside Kuaishou's short-video ecosystem rather than as a standalone demo. Generation is a feature in a creator funnel with hundreds of millions of viewers: successive model generations improved motion and physical consistency, monetized through creator subscriptions and an API.

    Key Insight

    The moat is distribution, not the diffusion transformer. Rivals ship impressive demos; Kling ships to an audience that immediately posts, monetizes, and feeds engagement data back — which is why it out-earns peers with comparable models.

    Groq LPU

    Hardware & Robotics
    Groq·Inference Silicon
    View product

    Tech Stack

    SRAM-only memory
    Static compiler scheduling
    Deterministic execution
    No HBM
    GroqCloud API
    3M+ developers

    How it works

    Inference silicon that throws out the GPU playbook: no HBM and no dynamic scheduling. Model weights live in on-chip SRAM spread across many chips, and a compiler schedules every cycle statically — so latency is deterministic and single-request token speed is extreme. Sold as the GroqCloud API rather than as boards.

    Key Insight

    Groq bet that inference is a compiler problem, not a memory-bandwidth problem. Determinism turns hardware into a predictable pipeline — and the 3M+ developers on GroqCloud, not chip specs, are what cleared the catalog's hardware evidence bar.

    Tesla FSD

    Hardware & Robotics
    Tesla·End-to-End Neural Driving
    View product

    Tech Stack

    Vision-only sensing
    End-to-end neural net (v12+)
    Fleet data engine
    Shadow mode
    HW4 in-car compute
    OTA deployment

    How it works

    From v12, FSD replaced hundreds of thousands of lines of hand-written control heuristics with a single end-to-end network from camera pixels to steering and pedals. It trains on fleet video curated by a data engine, validates candidate models silently in shadow mode against human drivers, and ships over-the-air to millions of cars.

    Key Insight

    The fleet is the lab: every mile generates training data and every OTA push is a deployment. The contested bet is vision-only — trading sensor redundancy for a data flywheel that no lidar-equipped fleet can match in scale. This is the clearest industrial preview of the embodiment wave.

    More teardowns coming soon

    We're analyzing how every product in our database uses AI under the hood