#OpenAgentHackathon - Twitter Hashtag

2 days ago

The villains never stood a chance. 🏆 Congrats to the 12 Open Agent Hackathon winners: Nightlist, SkyLedger, Fieldwork, GreenTrace, Proof of Stage, Proven, FocoCerto, Arca, Portorium, Athena, Flusso & Tradeflow. 5,964 builders. 138 countries. 549 agents. #OpenAgentHackathon

0

54

Jennycruzy

@jennyoliver57

3 days ago

I placed 3rd in Track 1 at the #OpenAgentHackathon by @GenAI_Now . Thank you to the judges and the team behind it. What I built Fieldwork: You ask a question about a neighborhood in plain English, like "which parts of Montgomery have high asthma rates and a lot of housing code violations?" No single website has that answer. The health numbers are on the CDC site, the violations are on the city's site, and the neighborhood map is somewhere else. Fieldwork goes and finds those sources by itself, connects them through @Zetaris, checks that they really line up, and answers on a map. Every number links back to the real records it came from. If the data doesn't exist, it says so instead of guessing. Try it: https://t.co/dCYJGcXMG3 Now that the leaderboard scores are out, here is what I learnt: I finished 1 point behind 1st place in my track. My Product & UX score was the best in the track. My build quality and video scored as high as the winners'. The whole gap was in one place: Impact. Judges asked who exactly this helps, and what that person does differently tomorrow because of it. I aimed too wide. Fieldwork was built for city staff, health analysts, reporters and residents, and a tool for everyone is hard to judge as vital to anyone. Finding the truth in scattered data is half the job. The other half is turning it into the next move for one specific person. What I will do next time: Keep everything that scored well, and add what was missing: - One person I am building for, with a real cost when the problem goes unsolved - An answer they can act on, not just read - Something they come back to every day. Thank you again to the judges and team for publishing these scores because It helps me know where to improve and what to keep doing. Lets build more!!

jennyoliver57's tweet photo. I placed 3rd in Track 1 at the #OpenAgentHackathon by @GenAI_Now .

Thank you to the judges and the team behind it.

What I built
Fieldwork: You ask a question about a neighborhood in plain English, like "which parts of Montgomery have high asthma rates and a lot of housing code violations?"

No single website has that answer. The health numbers are on the CDC site, the violations are on the city's site, and the neighborhood map is somewhere else. Fieldwork goes and finds those sources by itself, connects them through @Zetaris, checks that they really line up, and answers on a map. Every number links back to the real records it came from. If the data doesn't exist, it says so instead of guessing.

Try it: https://t.co/dCYJGcXMG3

Now that the leaderboard scores are out, here is what I learnt:

I finished 1 point behind 1st place in my track. My Product & UX score was the best in the track. My build quality and video scored as high as the winners'.

The whole gap was in one place:

Impact. Judges asked who exactly this helps, and what that person does differently tomorrow because of it.

I aimed too wide. Fieldwork was built for city staff, health analysts, reporters and residents, and a tool for everyone is hard to judge as vital to anyone.

Finding the truth in scattered data is half the job. The other half is turning it into the next move for one specific person.

What I will do next time:

Keep everything that scored well, and add what was missing:
- One person I am building for, with a real cost when the problem goes unsolved
- An answer they can act on, not just read
- Something they come back to every day.

Thank you again to the judges and team for publishing these scores because It helps me know where to improve and what to keep doing.

Lets build more!!

Jennycruzy

@jennyoliver57

6 days ago

I asked an AI agent a question no single dataset can answer. It went and found the data itself. This is Fieldwork, my solo build for the #OpenAgentHackathon by @GenAI_Now . My track Track 1: Solving Fragmented Intelligence. The problem: Ask "which Montgomery neighborhoods have high asthma rates and a lot of housing code violations?" The asthma data is on the CDC site, the violations are on the city portal, the neighborhood map is from the Census. No one has that table. What Fieldwork does: You type the question in plain English. It searches public data portals live, connects the sources through @Zetaris without copying any data, checks they really match up, and answers on a map. Every number links back to the real rows it came from. If the data doesn't exist, it says so. In the demo I also point it at Louisville, a city it had never seen. It found the sources and answered in 92 seconds. What I enjoyed most: Seeing my work help the sponsor directly. While building, I hit data sources on Zetaris's own "known to fail" list, so I fixed them and sent pull requests. Both were merged by Zetaris's CTO: - PR #36: Singapore air quality, now working https://t.co/mnR60by4B8… - PR #39: CDC health data + Census maps, added as new sources https://t.co/mnR60by4B8… "Super helpful… always great to add another confirmed data source into the set of known to work." After the hackathon I kept going. Using the @cursor_ai agent, PR #43 (merged this week) took the known-to-fail list from 6 sources to 1 https://t.co/mnR60by4B8… I also had the Cursor agent check Fieldwork's answers. It used only Zetaris and never saw Fieldwork's answers. On 5 of my questions it got the same numbers on 4, and the same top 5 on the last, with a few counts off by 1 or 2. Tools that made the biggest difference: - Zetaris: one SQL query across live CDC data and my own database, with every answer saved back as a reusable view - Meterless engines: H-MEM so the agent remembers which sources worked, World Model to track each run, Scout Intent to block poisoned source text - @nvidia Nemotron: writes the SQL and the answers What I learned, and what I'd do differently: Testing beats demos. On 25 real questions my first full run passed 12. The final run passed 25 of 25, about 47 seconds per answer. The same AI model on its own answered 0 of 20; Fieldwork answered 18. Next time I'd start that full test on day one, not day three. Thank you @GenAI_Now , @Zetaris and Meterless for a great week📌🔥 Try it: https://t.co/dCYJGcXMG3 Code: https://t.co/TU83EXHQUz…

jennyoliver57's tweet photo. I asked an AI agent a question no single dataset can answer. It went and found the data itself.

This is Fieldwork, my solo build for the #OpenAgentHackathon by @GenAI_Now .

My track

Track 1: Solving Fragmented Intelligence.

The problem:

Ask "which Montgomery neighborhoods have high asthma rates and a lot of housing code violations?" The asthma data is on the CDC site, the violations are on the city portal, the neighborhood map is from the Census. No one has that table.

What Fieldwork does:

You type the question in plain English. It searches public data portals live, connects the sources through @Zetaris without copying any data, checks they really match up, and answers on a map. Every number links back to the real rows it came from. If the data doesn't exist, it says so.

In the demo I also point it at Louisville, a city it had never seen. It found the sources and answered in 92 seconds.

What I enjoyed most:

Seeing my work help the sponsor directly. While building, I hit data sources on Zetaris's own "known to fail" list, so I fixed them and sent pull requests. Both were merged by Zetaris's CTO:
- PR #36: Singapore air quality, now working
https://t.co/mnR60by4B8…
- PR #39: CDC health data + Census maps, added as new sources
https://t.co/mnR60by4B8…

"Super helpful… always great to add another confirmed data source into the set of known to work."

After the hackathon I kept going. Using the @cursor_ai agent, PR #43 (merged this week) took the known-to-fail list from 6 sources to 1
https://t.co/mnR60by4B8…

I also had the Cursor agent check Fieldwork's answers. It used only Zetaris and never saw Fieldwork's answers. On 5 of my questions it got the same numbers on 4, and the same top 5 on the last, with a few counts off by 1 or 2.

Tools that made the biggest difference:

- Zetaris: one SQL query across live CDC data and my own database, with every answer saved back as a reusable view
- Meterless engines: H-MEM so the agent remembers which sources worked, World Model to track each run, Scout Intent to block poisoned source text
- @nvidia Nemotron: writes the SQL and the answers

What I learned, and what I'd do differently:

Testing beats demos. On 25 real questions my first full run passed 12. The final run passed 25 of 25, about 47 seconds per answer. The same AI model on its own answered 0 of 20; Fieldwork answered 18. Next time I'd start that full test on day one, not day three.

Thank you @GenAI_Now , @Zetaris and Meterless for a great week📌🔥

Try it: https://t.co/dCYJGcXMG3
Code: https://t.co/TU83EXHQUz…

7

24

7

3

2K

13

54

1

9

1K

Jennycruzy

@jennyoliver57

6 days ago

I asked an AI agent a question no single dataset can answer. It went and found the data itself. This is Fieldwork, my solo build for the #OpenAgentHackathon by @GenAI_Now . My track Track 1: Solving Fragmented Intelligence. The problem: Ask "which Montgomery neighborhoods have high asthma rates and a lot of housing code violations?" The asthma data is on the CDC site, the violations are on the city portal, the neighborhood map is from the Census. No one has that table. What Fieldwork does: You type the question in plain English. It searches public data portals live, connects the sources through @Zetaris without copying any data, checks they really match up, and answers on a map. Every number links back to the real rows it came from. If the data doesn't exist, it says so. In the demo I also point it at Louisville, a city it had never seen. It found the sources and answered in 92 seconds. What I enjoyed most: Seeing my work help the sponsor directly. While building, I hit data sources on Zetaris's own "known to fail" list, so I fixed them and sent pull requests. Both were merged by Zetaris's CTO: - PR #36: Singapore air quality, now working https://t.co/mnR60by4B8… - PR #39: CDC health data + Census maps, added as new sources https://t.co/mnR60by4B8… "Super helpful… always great to add another confirmed data source into the set of known to work." After the hackathon I kept going. Using the @cursor_ai agent, PR #43 (merged this week) took the known-to-fail list from 6 sources to 1 https://t.co/mnR60by4B8… I also had the Cursor agent check Fieldwork's answers. It used only Zetaris and never saw Fieldwork's answers. On 5 of my questions it got the same numbers on 4, and the same top 5 on the last, with a few counts off by 1 or 2. Tools that made the biggest difference: - Zetaris: one SQL query across live CDC data and my own database, with every answer saved back as a reusable view - Meterless engines: H-MEM so the agent remembers which sources worked, World Model to track each run, Scout Intent to block poisoned source text - @nvidia Nemotron: writes the SQL and the answers What I learned, and what I'd do differently: Testing beats demos. On 25 real questions my first full run passed 12. The final run passed 25 of 25, about 47 seconds per answer. The same AI model on its own answered 0 of 20; Fieldwork answered 18. Next time I'd start that full test on day one, not day three. Thank you @GenAI_Now , @Zetaris and Meterless for a great week📌🔥 Try it: https://t.co/dCYJGcXMG3 Code: https://t.co/TU83EXHQUz…

Jennycruzy

@jennyoliver57

16 days ago

Montgomery, Alabama's neighborhood data is spread out. Code violations and 311 requests are with the city, health data with the CDC, and boundaries with the Census Bureau. None of it lines up. So I built Fieldwork for the @GenAI_Now Hackathon Track 1; Solving Fragmented Intelligence Ask a question in plain English. Seven agents find the right public data, join it in @Zetaris , and check every number before you see it. Click any figure to see the record behind it. What it found: 10 of Montgomery's 71 tracts hold 48% of 2025 code violations. It scored 23 of 25 on a fixed test, and every run is published. The same model without the agents answered 0 of 20. Built with @nvidia Nemotron, Zetaris and Meterless. Try it: https://t.co/es3Tbs3QFJ

2

13

1

2

3K

7

24

7

3

2K

Cipoklean

@David63694631

16 days ago

CI went red. Why? RedRun is my solo build for the #OpenAgentHackathon: an agent that reads the failed run, the diff and dependency releases, then gives one cited cause plus a next step. Zetaris + Meterless Nightshift + Nemotron. 14/15 real failures.

0

2

0

158

bright /on @_ymai

30 days ago

Excited to be a Regional Ambassador for the Open Agent Hackathon 2026! 🚀 🗓️ Oct 22-27 | 🌐 Online | 💰 Up to $20K in prizes Build AI agents that do real work. Free, solo or in teams of up to 5. 🔗 https://t.co/e8JfemHnMJ #OpenAgentHackathon #AIAgents @GenAI_Now

1

17

5

2

332

Top Tweets for #OpenAgentHackathon

Last Seen Hashtags on Sotwe

Trends for you

Most Popular Users