A study built on Harumi platform shows that, once your forecast is reasonably accurate, how you act on it can matter more than a small improvement in accuracy.
#optimization#forecast#ai
Fresh polish from the team ๐ฅ๏ธ
The Scheduled badge on the project list is now clickable. One click shows the next run, how often it repeats and who gets the email, plus a shortcut to edit.
Finished one-time runs drop off, so only what's coming shows.
@andrehkoga
When people ask us whether we're competitors of big AI companies such as Anthropic or OpenAI, I always answer the same thing: We're working in tandem with these cutting edge models, not against;
Our platform harnesses their power, using a solid infra to let users run whatever optimization they want.
If models plateau or slow down in terms of improvements, that doesn't change our goals to become the lowest friction platform for OR and PO.
If models become strong enough to solve complex scheduling, routing or whichever kind of problem you got, then our platform will be even better at harnessing their capabilities.
Check what we're building at https://t.co/fMcRcIq4cj
@harumi_ai_
Iโll be at the @INFORMS Annual Meeting in San Francisco, Nov 1โ4!
If youโre working on operations research/optimization, Iโd love to connect & exchange ideas!
Who else is going? ๐
#informs#OR#Optimization
3/ Runs from the command line now show live status with elapsed time, and importing a folder previews exactly what gets uploaded while keeping private files like passwords and keys out.
What should the team ship next?
1/ Fresh drop from the Harumi team ๐ฅ
Harumi API keys now connect your own systems to your optimization results. Pull the latest outputs into your dashboards or start a run from your own tools. Each key only reaches the projects you pick.
@andrehkoga
2/ Security got a boost too.
Two-factor authentication is now in account settings, and sensitive actions like deleting an organization ask for a fresh authenticator code. A key also stops working the moment someone loses access.
Fresh from our team: tighter security controls ๐
Owners can now require extra verification across the whole organization, for admins and owners or for everyone. It stays off until turned on. Once it is, role changes, project deletion and share links ask for it.
3/ Made a bad change to a file? Pick an earlier version from its history and restore it in a click. Nothing is lost, and the restore can be undone too.
The team is building this in public. Which of these would you use first?
1/ Three quality-of-life upgrades just landed from our team ๐งต
Uploaded files now preview right in place. Images show inline, and text, PDFs and spreadsheets open in a viewer, so there's no downloading just to check you grabbed the right file.
2/ Usage is now shown as simple weekly and monthly percentages, plus a dollar cap you set for any extra spend.
See how much room is left and know the most a busy month can cost before it happens.
1/ Two fresh updates from the Harumi team ๐งต
Set your company's rules and preferences once, and every project picks them up.
Shared standards live at the organization level, personal ones stay yours, so a new project starts out already knowing how you work.
@andrehkoga, my brother & CTO, shipped scenario comparison at @harumi_ai_ this week!
Explore up to four scenarios side by side to see how inputs affect operational efficiency.
Compare results across Gurobi, CPLEX, and HiGHS to find the best solver for your problem.
#operationsresearch #math #ai
Have you checked out the first AI-native development platform for operations research teams?
Designed for both in-house OR teams and consulting firms, Harumi helps you build, run, and deploy optimization systems with AI.
1/ Fresh upgrades from the @harumi_ai_ team ๐งต
Long analyses in chat now run all the way through: no more big jobs stopping halfway. Ask Harumi for the deep dive and it sees it to the end.