Comment by smallmartial

Back to stories Open in a window
Comment

Comment by smallmartial

Hey HN,

We've been working on Reelsy for the past few months and wanted to share what we've learned building a multi-agent AI system for video content creation.

The Problem

Creating short-form video content (YouTube Shorts, TikTok, Reels) at scale is brutal. A single 60-second video costs $500+ with freelancers and takes 3-5 hours. For creators who need to post daily, this is unsustainable.

The bigger technical challenge: AI-generated images have a consistency problem. Ask any image model to generate "the same character" across 15 scenes, and you'll get 15 different-looking people.

Our Approach: Multi-Agent Architecture

Instead of throwing one LLM at the problem, we built a 5-agent pipeline inspired by actual film production:

Director Agent - Breaks down the story concept into a shot list Scriptwriter Agent - Writes dialogue and narration for each scene Character Designer Agent - Creates reference images and locks character identity Cinematographer Agent - Determines camera angles, lighting, composition Hook Generator Agent - Optimizes the first 3 seconds for each platform The agents communicate through structured outputs and can iterate on each other's work. We found this beats a single mega-prompt approach by ~40% on our internal quality benchmarks.

Character Consistency Solution

We use Gemini 2.5 Flash ("Nano Banana" on LMArena) with reference image anchoring. The Character Designer creates a canonical reference image, then every subsequent scene generation includes this reference with specific instructions to maintain identity.

Current results: 85%+ consistency across 15-20 scene generations. Not perfect, but usable for most content types.

Multi-agent coordination is harder than it looks. Race conditions, agent disagreements, and context window limits are real problems.

Character consistency is still the hardest unsolved problem in AI video. Reference anchoring helps but isn't bulletproof.

Platform-specific optimization matters more than raw quality. A slightly lower quality video with proper hooks outperforms beautiful content with weak openings.

Current Pricing

~$0.70 per video (15-20 scenes, voiceover, music). We're not trying to undercut human editors on quality, but for high-volume content needs, this makes previously impossible workflows viable.

Would love feedback from the HN community, especially on the multi-agent architecture. Happy to answer questions about our implementation choices.

Replies smallmartial · 2026-01-04
Open on HN
Loading the discussion…

Domain filters

Stories from these domains are hidden from every list. Subdomains match too: blocking substack.com also hides danluu.substack.com.

    About YAVCHN

    YAVCHN is a reader for Hacker News and Lobsters, with articles and discussions in separate windows or Classic pages.

    Created by Paul Parks and built with PUDL.

    YAVCHN source code on GitHub

    Help

    Keyboard

    j / k
    Move down and up the story list. The arrow keys scroll whatever has focus.
    Enter
    Read the marked story in the article reader.
    ]
    Read the next story in the same article-reader applet. Back returns to the previous story.
    p
    Pin or unpin the marked story, which keeps it in Pinned.
    n / N
    Move to the next or previous top-level comment in the window in front.
    c
    Collapse or expand that comment.
    f
    Hide or show the story list.
    Esc
    Close a menu or this help.
    Access key m
    Go to the menu bar. Most browsers take it with Alt on Windows and Linux, and Safari with Control and Option.
    ?
    Show this help.

    Windows

    Each story opens in a window holding its article above its discussion; drag the bar between them to share the room differently. A window can be moved by its title bar, resized from any edge, snapped to a half or a corner by dragging it there, maximised, or minimised to the bar at the foot of the page. Use Window > New reader window to open an empty reader, or a story row's Open in new reader window button to compare articles. Docked readers keep their articles when you select another story from the sidebar. Minimized readers can be restored and reused for their site. A window's Next story link reads on down the list in the same window.

    A link in a comment or an article to another Hacker News or Lobsters thread opens that thread in a window too. A link to a single HN comment opens the comment above its replies.

    While a story's window is in front, the Story and Discussion menus in the menu bar hold its commands: pinning, Next story, sorting, collapsing every thread, jumping to the first new comment. Each window also remembers where you were in its article and discussion, so a reload, or Back to a story that Next took you past, finds your place again. Closing a window forgets it.

    The whole arrangement lives in the address, so a bookmark or a shared link brings it back, and Back undoes the last change. Moving between Hacker News, Lobsters, their lists, Pinned and Find changes only the list, and leaves the windows open.

    The list

    The pin at the start of a row keeps the story in Pinned, and the cross at its end hides it. Pinned can be narrowed by words in the title, site or author, by source, and to the stories you haven't opened yet, and ordered by when you pinned them, by points or by comments; the filters are part of the address, so a filtered view can be bookmarked. Scroll past the end of the list to load more. Domain filters, in the View menu, hide every story from a site.

    Browsing view

    View > Windowed and View > Classic select the browsing view and save your default in this browser. Window view reuses a reader for each feed. Classic view opens stories and applets as pages. Open as a page and Open in a window are one-off actions that do not change your saved default. Direct page links always open as pages.

    Applets

    The Applets menu in the menu bar holds three tools, each a window of its own. Replies to me takes your Hacker News user name and lists the replies to your last thirty comments and stories, checking again every three minutes while it is open, and marking what is new since you last marked them read. Look up a user opens a profile on Hacker News or Lobsters, with their submissions and recent comments, as a commenter's name in any discussion does; the bar at the top of a profile looks up someone else in the same window, and Back returns to the one before. Who is hiring? filters the posts of HN's monthly hiring threads by the words you type.

    They read only what the sites publish to everyone, so none of them asks for a login, and your user name stays in this browser.

    Find

    Find takes any link and lists every time it was submitted to Hacker News and Lobsters, so you can read each discussion of it.

    About

    YAVCHN never sees your Hacker News or Lobsters login. The discussion is fetched from each site's public API; to vote or reply, follow the link above the discussion, or the arrow beside a comment, to the source's own site. Pins, hidden stories, filters and layout are kept in this browser only.

    Open source: github.com/paulmooreparks/yavchn. Built with PUDL.