Self-Promotion
Show off what you're working on
Start new thread

What are you building? Drop your AI product below

by•

Doing one of these because I get more from reading what others are building than from any newsletter.

I'll start.

Building AI Hive, an enterprise AI agent platform that helps mid-market and enterprise teams get from AI pilot to production in weeks instead of quarters. The hard problem we keep solving: compliance, model flexibility, and the lack of in-house AI talent on the customer side.

Product Hunt page:

Your turn. Drop:

- What you're building

- Who it's for

- The hardest part you're solving right now

Will read everything and upvote what resonates.

4K views

Add a comment

Replies

Best

Oops, I didn't catch that this was a thread about "AI products." My apologies, and remove my comment if you want. I didn't mean to post in a non-relevant thread.

 it's okay, no worry since you can freely share your launching product here

SoloDev here (doing this outside of my 9-5); I've decided to start shipping instead of remaining in a state of perpetual build. I'm shipping the tool I use most in my own work: Img-forge. but


is the platform UI; but Easiest way to get your agent working with it is via mcp connection . I am currently registering the mcp server in applicable registries.


Once a user base is built up a tad I will be able to unlock OpenAI access and start working on vid with Veo and other options. I'll be working on some prompting guides as well. Docs are at:

Current known issues being addressed today:

  • Access token 15 minutes. No refresh token is issued — the `/token` grant only supports `authorization_code`; re-run the PKCE flow to get a new access token. I will be increasing that token duration.

  • Increasing the 500/mo Flat-tier cap to accommodate Pro and Agency users

    I'm doing this while working FT. I want to help people ship. Its for builder of all skills, and the hardest part is doing this solo.

 Thanks for the transparency here, posting known issues (token duration, tier caps) openly while still shipping is rare and honestly builds more trust than a polished launch page would.

Solo dev shipping while working full time is a grind I respect a lot. The MCP connection approach is smart, meeting builders where their agents already live instead of asking them to learn a new UI first.

Followed and upvoted, good luck getting that user base rolling. Would appreciate you checking out AI Hive too if you have a sec, automation flows for productivity, happy to upvote/follow back:

Building FounderFlow — an AI Executive Chief of Staff for founders running more than one business at once. It watches what's happening across all your businesses, surfaces the 1-2 things that actually need your attention today, and grades its own confidence (Verified / Very Likely / Needs Review / Monitor Only) instead of pretending to be sure when it isn't. Built it for myself after 30 years in org development and running a software company, a care home, and a cafe simultaneously. Would love feedback from anyone juggling multiple ventures:

 Thanks for sharing this, building it for yourself after running a software company, a care home, and a cafe at once says everything about why this needed to exist. Most "AI chief of staff" tools feel like they're built by people who've never actually juggled that much.

The confidence grading (Verified/Very Likely/Needs Review/Monitor Only) is the detail that stands out most, most tools pretend certainty they don't have and that erodes trust fast. That kind of honesty is rare.

Followed and upvoted, would love updates as you build this out further. If you have a moment, please check out AI Hive too, automation flows for smoother workflows, an upvote/follow back would be appreciated:

 Thanks Nolan, really appreciate you reading it closely. The confidence grading came straight from beta feedback, people have been burned by tools that state guesses as fact, so "Verified" vs "Monitor Only" ended up mattering more than any feature. I'll take a genuine look at AI Hive.

 Thanks for engaging so openly on this, genuinely curious to see where you take it. The beta feedback point about people getting burned by tools stating guesses as fact really resonates, we've had similar conversations internally about how easy it is to overpromise confidence when the model just doesn't know. Good luck building it out further.

 Thanks Nolan, appreciate you following the thread. It's still early days so there's plenty I haven't gotten right yet, but the confidence grading is the one piece I'd never ship without now.

What I'm building:

Who's it for: Students, small business owners, project managers

Hardest part right now: Having AI wired in to automatically create lists, cards and fill descriptions through prompt via helper.

Free to try:

 Thanks for sharing, Kanban boards for students and small biz owners is an underserved niche, most tools out there are built for engineering teams and it shows.

AI auto-creating lists/cards/descriptions from a prompt is the right instinct, manual board setup is the single biggest reason I've seen people abandon these tools in week one. Tried the free version and the interface felt light, which honestly is a plus here.

What we’re building

— a simple, fast, and lightweight task management platform for planning, tracking, and completing work without unnecessary complexity.

Who it’s for

Freelancers, startups, small teams, and growing companies that find tools like Jira, ClickUp, or Trello too complicated, heavy, or distracting.

The hardest part we’re solving right now

Creating a task management experience that stays simple as teams and projects grow. We’re working on adding useful features without turning into another overloaded productivity platform.

We’d love to hear your feedback: what feature would make more useful for you?

 Thanks for building something for this exact problem, Jira and ClickUp fatigue is very real once you're not managing a huge team anymore. Simplicity as the actual selling point instead of "more features" is refreshing.

The hard part you called out, staying simple as teams grow, is the one every tool eventually messes up. Feature request from me: some kind of "complexity budget" warning before you add a new field or view, so it forces you to stay honest about scope creep.

Followed and gave it an upvote. If you don't mind, would love your eyes on too, we're building automation flows for smoother team ops, an upvote/follow back would be great:

What I'm building:
PromptVault — a five-volume AI prompt series built on peer-reviewed research. Not templates. Every prompt starts with validated data, gets an editorial filter applied, and earns its place or gets cut

Who it's for:
Founders, operators, marketers, and e-commerce brands who have bought AI prompt packs before, found them generic, and want the version built to a different standard

The hardest part I'm solving right now:
Most AI prompt products are like not differentiable from each other — same frameworks, same formats, same lack of research behind any of it. The real problem isn't that people need more prompts. It that they need fewer, better ones they can actually trust. That's what PromptVault is
My bigger plan is to launch a community of prompt engineers, testers and users, where people can come to get access to the most widely tested prompts around the globe.

 Really like this framing, "people don't need more prompts, they need fewer better ones they can trust" is such a clean way to call out how saturated and generic that space has gotten. Building on peer reviewed data instead of just templates feels like the right way to actually earn that trust back.

Hope PromptVault grows into that community you're picturing, sounds like a real gap in a market that's mostly noise right now.

Hi there. We're Fluent Terrain and we're building GlacialBooks, an automated bookkeeping platform that auto categorizes transactions, automates month end close, and reconciliation.

For: Small, new, service-based businesses, freelancers, agencies and contractors.

The hardest part we're solving is cash deposit transaction categorization.

View our product page here:


Try our product beta here:


We're launching July 31.

 Thanks for sharing GlacialBooks, cash deposit categorization is such an annoyingly specific problem, most bookkeeping tools just kind of hand wave it and dump it in "uncategorized." Freelancers and small service businesses genuinely need someone solving that instead of ignoring it.

Hope the July 31 launch goes smoothly and the product keeps getting sharper from there, upvoted and following to see how it lands. If you get a chance, would really appreciate you checking out AI Hive too, it's built around helping teams set up automation flows so work moves without all the manual back and forth, an upvote or follow would mean a lot:

What we're building
We're doing to immigration what TurboTax did to taxes.

Today, moving abroad means navigating dozens of government websites, lawyers, appointments, documents, and unanswered questions. We're replacing that chaos with a single AI-powered platform that finds your best path to legal residency in under 3 minutes, then matches you with a licensed immigration lawyer and coordinates every step until you're legally settled. Making it effortless to live, work, move, and thrive anywhere.

Everyone else helps you understand the process. We're building the infrastructure that gets it done.

Who it's for

People who want to build a life in a new country without spending months figuring everything out on their own. We currently help migrants relocating to Spain, with plans to expand internationally.

The hardest part we're solving right now

Turning one of the world's most fragmented, human-driven processes into a scalable product without losing the trust, personalization, and legal expertise that life-changing immigration decisions require.

You can try our beta here -->

 The TurboTax analogy actually nails it. Immigration is one of those processes that's completely broken for the end user but nobody's really fixed it at the infrastructure level, just added more layers of PDFs and lawyer portals on top. The "find your path in 3 minutes" part is what I'd want to see hold up in practice, especially for edge cases like dual citizenship or work visa conversions.

Rooting for this one. Will follow to see how the Spain rollout goes. If you get a chance, would love your take on too, we're building automation flows for enterprise teams and always keen to hear from founders solving hard workflow problems:

Hey, I'm building ListenDock: Listen to your PDFs and get them explained with audio.

 Turning PDFs into audio explanations is such a practical idea, so many of us have a backlog of documents we'll never actually sit down and read but would happily listen to during a commute or workout.
Hope ListenDock keeps growing, upvoted and following already. Would love your feedback on AI Hive too if you're open to it:

What I’m building: Clarity — a browser/PWA meeting recorder that turns a real call into a summary, decisions, a searchable transcript, and specific action items with owners, dates, and blockers. No app install and no meeting bot joins.

Who it’s for: freelancers, consultants, indie founders, and agency PMs who run client calls, 1:1s, or kickoffs.

Hardest part right now: getting people past signup and into one real recorded meeting, where the action items become concrete enough to replace rewriting notes afterward.

The one-time trial includes 60 hosted minutes:

 No app install and no meeting bot joining the call is such a relief honestly, half the meeting recorder tools out there feel intrusive before they even prove useful. Getting people from signup to one real recorded meeting where the notes actually replace manual rewriting is the right thing to focus on first.
Hope Clarity keeps making that trial moment count, upvoted and following.

@nolan_vu Thanks — that’s exactly the tradeoff I’m trying to solve. If you run a client or team call this week, would you try one real recording and tell me whether the decisions and action items are specific enough to use without rewriting them? The trial includes 60 hosted minutes:

First
Previous
•••
131415
•••
Next