




AgentForge
Build agents. Test them. Prove they work.
Links
Additional info
How was your experience building with Codex?
Building with Codex was fast, smooth, and highly productive. It helped us move from idea to implementation quickly, debug issues, improve UI/UX, and structure the project better. Codex felt like a coding partner that accelerated development while still keeping us in control of the final decisions.
Describe your experience using Loops House as the hackathon platform. What worked well, what challenges (if any) did you face, and what improvements would you like to see?
Loops House provided a clean and organized platform for managing the hackathon. The submission flow was simple, the event information was easy to follow, and the overall experience felt smooth. A useful improvement could be adding more real-time submission status updates and clearer visibility into judging timelines.
Tell us about your overall experience at Codex Community Hackathon Pune.
The overall experience was exciting, collaborative, and inspiring. It pushed us to build a practical product in a short time, learn faster, and think deeply about real-world use cases. The hackathon gave us a great environment to experiment, build, and present something meaningful with AI.
What could Codex Community improve to create a better experience for participants?
Codex Community could improve the participant experience by providing clearer timelines for each stage, more frequent updates during the hackathon, and detailed judging criteria in advance. It would also help to have more mentor check-ins, quick technical support channels, and a short feedback session after submissions so teams can learn and improve from the experience.
Team
4 members- DEOwner
Dev Sarangdhar
- MA
Manav Yadav
Prompt Engineer & Tester - PR
Pratham Shah
Backend - SO
Soham Dandawate
UI/UX
Overview
AgentForge is an AI agent prompt workbench built for developers, startups, support teams, and businesses that want to create reliable domain-specific AI agents. Users can generate agents from templates, customize prompts and instructions, test them on real-world tasks, evaluate responses with quality scores, improve weak prompts, export run history, and generate deployment keys for future integrations. It turns agent building into a complete workflow: create, test, evaluate, improve, and deploy.