manhattan-reasoning-gym is an open-source SDK and CLI tool for FPGA developers to perform local synthesis, place-and-route feedback, and run sandboxed agents on real silicon. It supports reinforcement learning workflows and hardware integration, making it suitable for advanced hardware design and experimentation.
manhattan-reasoning-gym is a Frameworks & SDKs product. It enables local synthesis, place-and-route feedback, and agent sandboxing for FPGA hardware design. It is built as an open-source project for FPGA developers and hardware engineers. manhattan-reasoning-gym is open source under the MIT license. It runs on the web and the command line.
It is developed by Barnard-PL-Labs, and the product first shipped in 2026. Development happens publicly on GitHub with 2 commits in the last 90 days. PulseGate's similarity index finds few close equivalents — manhattan-reasoning-gym occupies a relatively distinct niche. Key capabilities include FPGA synthesis, pnR feedback, and sandboxed agents.
Latest indexed changes and source events
manhattan-reasoning-gym discovered by the PulseGate indexer
Other apps tracked under the same category.
