Aman Sachan

@AmSach

Don't bother

voxelnav. Real-time semantic voxel mapping for ROS2 robots - 100ms latency on Jetson Nano

23

kvquant. Drop-in KV cache compressor for local LLM inference - Run 70B models on 8GB RAM

7

GhostPilot. GPS-Denied Drone Navigation with Agentic AI - VINS-Mono + Nav2 + LLM Mission Parser

4

bitforge. 🔥 Shrink any LLM to fit in your pocket - Auto-compress language models for microcontrollers

2

CMF-Infinity. Continous Meaning Field LLMs are the future

1

entropic-pegasus. Python

1

llm-foundry. Python

1

kernelweave. KernelWeave: self-compiling skill kernels for LLMs

1

WhisperBridge. Offline voice transcription CLI with speaker diarization - 3x faster than Whisper

1

Lingua. Google Tranlate Translation Agent which could translate a 600 page pdf in mere seconds and reconstruct it exactly the way you input.

1

mosaic-compress. Fast compression that beats bz2 in speed AND ratio - byte-plane entropy transform + zstd

1

NeuroShell. Convert plain English to safe shell commands using LLMs

1

datapop. Generate realistic synthetic tabular data with a single line of Python

1

SignalHop. Python

1

sentience. Sentience - Local AI Computer. Like Zo + Claude Code, 100% on your machine.

1

Agada. ur very own medicine assistant

1

nexus. OSINT engine

1
17
Apply