Launching today
Load Nova
An AI co-pilot and dashboard built for dispatcher speed
90 followers
An AI co-pilot and dashboard built for dispatcher speed
90 followers
One workspace for dispatchers. Less tab switching. Smarter freight. Load Nova is a Chrome side panel and Dashboard that lives next to your load board. Parse broker emails, calculate real RPM & profit, plan routes with live weather — end-to-end in under 3 minutes. No tab switching. Key features: Unified dispatch dashboard AI-assisted load planning Driver and load management Route & HOS awareness Email and workflow automation Real-time operational insights Work inside ChatGPT
Load Nova
Hey Product Hunt! 👋
I'm Anton, one of the founders of @Load Nova
The more we talked to dispatchers, the more we kept seeing the same thing: dozens of browser tabs, endless copy-pasting, and constant context switching.
Freight moves fast. And in this industry, speed isn't just convenience - it's money. Every minute spent looking for information is a minute not spent making decisions, helping drivers, or booking the next load.
We honestly believe dispatchers deserve better tools. In 2026, you shouldn't have to jump between 15–20 tabs just to understand what's happening with your fleet. The right information should be in one place, ready when you need it.
That's why we built our product.
Not to replace dispatchers but to help you work faster, make better decisions, reduce mistakes, and keep all the context in one workspace.
We're still building, still learning, and we're looking for early users who want to shape the product with us. If you're a dispatcher, carrier, or just curious about where AI in freight is going, we'd love to hear your thoughts.
Thanks for checking us out. We'll be here all day answering questions and collecting every piece of feedback. 🚚
RiteKit Company Logo API
@anton_mironov Congrats on the Load Nova launch, Andrii. A co-pilot and dashboard built around dispatcher speed is a sharp focus — that job lives or dies on how fast you can see and act, and most tools ignore it. Nicely aimed.
One idea while it's live: your PH launch is still editable, and a video in the gallery holds attention better than screenshots. So I made you one from your site, free, and it's whitelabel — no branding of mine on it, yours to post as your own: https://foxplug.com/v/load-nova-0580e8ef
I build FoxPlug — it turns your real product updates into videos, posts, and GIFs automatically: guided feature-release tours, animated GIFs, short posts for X, long-form for a blog, and full product walkthroughs. foxplug.com — building in public, loudly.
The broker-email parsing is the piece I'd dig into. Those emails are all over the place: plain text, forwarded chains with the real load buried three replies down, rate cons as PDF attachments. When we built an email-to-structured extractor, the accuracy cliff was always the quoted chains, where the model grabbed a stale rate from an earlier message in the thread. How does Load Nova decide which number in a back-and-forth is the current one, and does it read PDF rate cons or only the body?
Load Nova
@dipankar_sarkar Great question. We’ve seen the same issue: the hard part isn’t finding a rate, it’s knowing whether that rate is still current.
Load Nova treats the email thread as stateful, not as one big blob of text. Each new message is compared against the current load snapshot, so quoted/forwarded history doesn’t automatically overwrite newer info.
For rates, we look at where the number came from: the latest reply, a quoted message, or a rate con attachment. Clear updates replace the current value; ambiguous conflicts get flagged instead of silently changing the load.
And yes, when PDF rate cons are available, we read them too — not just the email body.
Treating the thread as stateful instead of one blob is the right model, that's the thing most extractors get wrong. The case that broke ours was recency not always meaning truth: a broker would forward the original rate con after sending a correction, so the newest message in the thread carried the stale number. Do you weight by message timestamp, or does the model reason about which message is actually the authoritative update?
Load Nova
@dipankar_sarkar Exactly. Timestamp alone is not enough.
A newer email can still carry an older rate con, forwarded history, or quoted context. So the way I think about it is not “latest message wins,” but “what is the role of this evidence?”
Is it a correction? A confirmation? A forwarded original doc? Or just historical context from the thread?
For fields like rate, we keep a resolved current state and only update it when the new evidence is strong enough. If it’s ambiguous, I’d rather flag the conflict than silently overwrite the load with a stale number.
That case you described is exactly why freight workflows need stateful reconciliation, not just extraction.
ChatWebby AI
The "15-20 tabs" pain point is so real for dispatch teams — having RPM, profit, and live-weather routing in one side panel next to the load board is a smart way to cut the context switching. I'm curious how the route planning handles multi-stop loads where HOS limits force a reset mid-route. Does it factor available driving hours into the suggested plan, or is that still on the dispatcher to layer in?
Load Nova
Thanks@zain_sheikh Yes that’s exactly what we’re solving. Our route planning takes available HOS, breaks, and resets into account, so dispatchers don’t have to calculate those constraints manually.
The goal is to replace tab-switching with a single, actionable view while keeping the dispatcher in control
Role-of-the-evidence instead of latest-wins is exactly the right frame. The part that broke ours was classifying the role itself: a genuine correction and a broker re-sending the original rate con both read as 'just confirming the rate is X', so the model kept treating a forwarded original as a fresh confirmation. Did you base that role call on email structure (reply depth, attachment vs inline) or on the language, and which one ended up more reliable?
Load Nova
Small note. Dispatching is moving from tabs to AI workspace 😁