AI research and projects
The research, projects, and explanations behind the people. Read the source, inspect verified code and models, and find a collaborator.
A source check confirms that a source supports the summary. It is not an independent replication or endorsement. Our methodology ↗
Voice-First Multimodal Agent Harnesses and Vibe Coding Workflows
Mckay Wrigley · Takeoff / Chatbot UIDeveloped real-time voice agent prototypes combining WebSocket audio streaming, low-latency function calling, and Cursor-driven rapid software development workflows.
WebSocket audio streams are sensitive to network packet jitter and browser audio permissions.
Context: OpenAI Realtime API, Cursor AI code generation, ElevenLabs streaming audio.
View missionWeb Cafe AI Client Knowledge Base & Agent Templates
Corbin Brown · Web Cafe AIEngineered reusable client delivery frameworks for syncing Notion, Google Drive, and Zendesk data into secure embedding indexes with role-based access control.
Incremental document sync requires webhook event listeners to capture upstream edits in real time.
Context: LangChain, ChromaDB, FastAPI, Next.js, Tailwind CSS.
View missionWriting An Interpreter and Compiler In Go
Thorsten Ball · Independent Author & Systems EngineerAuthored two landmark engineering books used by university curricula and tens of thousands of systems engineers worldwide to learn lexing, parsing, AST generation, and bytecode virtual machines.
Focuses on language implementation theory rather than modern JIT compiler optimizations.
Context: Go, lexer/parser construction, tree-walking interpreters, bytecode compilers.
View missionAlphaFold 2: Computational Structural Biology Prediction Engine Across 200M+ Proteins
Dr. John Jumper · Google DeepMindFirst author and technical architect of the open-source AlphaFold 2 computational system, predicting 3D coordinates of all cataloged proteins in the UniProt database with atomic precision.
Computational requirements for long sequence complexes (>2000 residues) require high-memory GPUs (A100/H100).
Context: JAX/Haiku, Multiple Sequence Alignment (MSA) representations, structural recycling iterations.
View missionCursor Speculative Code Editing & Shadow Workspace Architecture
Aman Sanger · Cursor / AnysphereArchitected Cursor's instant-apply and speculative editing engine, which runs isolated shadow syntax trees to predict multi-token keystrokes and merge large diffs into production codebases with zero visual latency.
Requires real-time AST parsing and language server protocol (LSP) indexing to ensure semantic validity of speculative edits.
Context: Custom fine-tuned small speculative draft models paired with frontier reasoning models (Claude 3.5 Sonnet, GPT-4o).
View missionFLUX.1: 12B Parameter Rectified Flow Transformer for High-Fidelity Image Generation
Dr. Robin Rombach · Black Forest LabsArchitected FLUX.1, an open-weights 12B parameter multimodal diffusion transformer utilizing rectified flow matching, achieving unmatched prompt adherence, typography generation, and photographic realism.
12B model footprint requires 24GB+ VRAM or 8-bit/4-bit quantization (NF4) for local inference on consumer hardware.
Context: 12B parameter multimodal Diffusion Transformer (MMDiT), rectified flow matching, rotary positional embeddings.
View missionllama.cpp: High-Performance C/C++ Foundation Model Inference Engine
Georgi Gerganov · ggml.aiEngineered a zero-dependency C/C++ LLM inference framework implementing 2-bit through 8-bit integer quantization (GGUF), ARM NEON, Apple Silicon Metal, CUDA, and AVX2 vector intrinsics, powering the modern local AI ecosystem.
Extreme low-bit quantizations (2-bit/3-bit) exhibit perplexity degradation on complex symbolic reasoning tasks.
Context: GGUF quantization format, Llama 2/3, Mistral, Qwen, DeepSeek; pure C/C++ runtime.
View missionMulti-File Codebase Indexing & Semantic Agentic Refactoring
Aman Sanger · Cursor / AnysphereEngineered local embedding and symbol graph pipelines capable of indexing hundred-thousand-line monorepos into contextual embeddings, enabling multi-step agentic software modifications.
Context budgets require strict retrieval pruning in repositories with dense circular dependencies.
Context: Custom tree-sitter chunking, semantic vector index, agentic terminal command execution loop.
View missionThe Mojo Programming Language for AI Hardware Kernel Development
Dr. Chris Lattner · ModularDesigned and built Mojo, a high-performance language that combines Python syntax and developer ergonomics with systems programming primitives, SIMD vectorization, and hardware-accelerated memory ownership.
Ecosystem packages and third-party library bindings are actively expanding relative to decades-old C++/CUDA tooling.
Context: Modular MAX Engine, Mojo compiler, heterogenous CPU/GPU execution pipelines.
View missionUnsloth: Fast and Memory-Efficient Open-Source LLM Fine-Tuning
Daniel Han · Unsloth AICreated an open-source library that rewrites PyTorch forward and backward passes into handwritten Triton kernels, supporting LoRA, QLoRA, and full fine-tuning with 0% accuracy loss while preventing out-of-memory errors.
Heavily optimized for NVIDIA Ampere, Ada Lovelace, and Hopper architectures; AMD ROCm support is emerging.
Context: Llama 3.1/3.2, Mistral NeMo, DeepSeek-R1-Distill, Gemma 2; 4-bit bitsandbytes and 16-bit mixed precision.
View missionvLLM: High-Throughput and Memory-Efficient LLM Serving Engine
Woosuk Kwon · vLLM / UC BerkeleyEngineered and open-sourced vLLM, an enterprise serving system providing continuous batching, chunked prefill, speculative decoding, and tensor parallelism, serving billions of production inference tokens daily worldwide.
Requires dedicated GPU clusters and NVLink interconnects for multi-node tensor-parallel deployments.
Context: CUDA/ROCm kernels, Triton backends, Ray distributed orchestration, vLLM OpenAI-compatible server.
View missionwhisper.cpp: High-Performance Automatic Speech Recognition in Pure C/C++
Georgi Gerganov · ggml.aiDeveloped a pure C/C++ port of OpenAI Whisper speech recognition with zero external dependencies, streaming audio support, WebAssembly compilation, and hardware acceleration on Apple Silicon and mobile devices.
Real-time streaming requires audio buffering and speech activity detection (VAD) to prevent hallucinated audio loops.
Context: OpenAI Whisper tiny/base/small/medium/large models, 16-bit float and 4-bit/5-bit/8-bit integer quantization.
View missionAutoResearch: Autonomous AI Research Agent Conducting Iterative ML Experiments
Andrej Karpathy · Eureka Labs / ex-Tesla, OpenAIAn autonomous multi-step research agent framework designed to conduct end-to-end machine learning experimentation without human intervention. The agent autonomously formulates algorithmic hypotheses, modifies PyTorch model architectures via AST rewriting, schedules multi-GPU training jobs, parses validation loss curves, and generates structured empirical research reports and pull requests.
Requires sandboxed execution environments to prevent arbitrary code execution risks during dynamic model definition; experiments can diverge if loss curves suffer from exploding gradients without bounded learning rate search spaces.
Context: PyTorch 2.4, LLM function calling, Automated Evaluation Loop, Python AST manipulation, TensorBoard metric parsing.
View missionllm.c: Large Language Model Training in Pure C/CUDA Without Dependencies
Andrej Karpathy · Eureka Labs / ex-Tesla, OpenAIA zero-dependency, pure C and CUDA implementation of GPT-2 and modern transformer pre-training. Eliminates the 245MB+ PyTorch/Python runtime overhead by directly implementing forward passes, backward passes, AdamW optimizers, and fused CUDA kernels in a single clean compilation unit, achieving near-parity with PyTorch cuDNN speed.
Lower level of architectural expressiveness compared to dynamic Python autograd graphs; adding novel attention variations requires manual mathematical derivation and custom CUDA kernel writing.
Context: Pure C99, CUDA 12, FP32 and BF16 mixed-precision kernels, fused attention, multi-GPU MPI execution.
View missionAutomated Code Review and PR Triage AI Assistant
Michael Ballard · SlashdevBuilt a GitHub Action and webhook bot that analyzes incoming pull requests, identifies security vulnerabilities and formatting anti-patterns, and generates unit test coverage proposals automatically.
Large pull requests exceeding 2,000 lines require chunking across logical Git commits to fit within prompt context ceilings.
Context: Claude 3.5 Sonnet, GitHub Octokit API, AST tree-sitter parsers.
View missionAutomated Content Repurposing Engine for Long-Form Media
Paul Macko · Deplatformable / Macko MediaBuilt an end-to-end automated workflow that ingests audio podcasts, generates precise Whisper transcripts, identifies viral hooks using LLM evaluation, and renders styled short-form video clips automatically.
Multi-speaker audio with heavy background music can lead to transcript misalignment without voice separation preprocessing.
Context: Whisper API, Claude 3.5 Sonnet, FFmpeg command automation, Python script runner.
View mission