human-machine symbiosis // supercoordination // physical AI // XR agents // computational cognition
browse 190+ repos · jump to the good ones ·
i'm Alif ! I keep coming back to the same problem: most software treats people as operators sitting in front of a screen, but the interesting work happens in rooms, in bodies, between people, across time. I want to build the interfaces for that. I massively support open source software, and am constantly experimenting and expanding my skills.
so in practice I end up somewhere between XR, brain-computer interfaces, multi-agent systems, and modular robotics.
I studied for minors in robotics && math, with my dual degree in Computer Science + Business Intelligence && Data Analytics with an undergrad at Clarkson University from 2019-2023 in the US, and at DTU in Denmark, and... I wanted to build new organizations, technologies, startups, philosophies, and scientific fields.
now I'm an affiliated researcher at the MIT Media Lab, Fluid Interfaces group for the last few years doing gnarly interdisciplinary and fundamental work.
I did MIT Reality Hack every year from 2022 to 2026 (fellow hackers reach out 2 me for opportunities!!), and now run Superintelligent Group AKA ΣIG.
before ΣIG, I worked through Halcyox, my distributed advanced HCI R&D company which eventually morphed into everything. Some of these repos are polished... some are messy, as a perfectionist I now try to build in public as much as I can, I know all too well that learning in a (vacuum) produces unproductive outcomes...
right now I advise engineering & research at Doppel Labs and Wallace. I’m also collaborating with Augmentiv Labs on a scientific reproducibility stack that uses smartglasses to make lab procedures verifiable and repeatable. from nov 2024 to aug 2025 I was the first non-founding engineer at Fabublox, an MIT Nanotechnology collaboration, simulation, and process design spin-out startup where I upgraded their process editor, created their version control system, and built out a full-stack AI system called PANDA (Process && Nanofab Design Assistant) to help Fab Services be more intuitive to access across facilities, among other awesome things (like accelerating the entire dev team w/ Vite && Cursor && Claude).
At Clarkson from 2019-2023, I conducted research at the Terascale All-Sensing Research Studio (TARS) under Sean and Natasha Banerjee in XR, robotics, CV, MLK, and AI. During u.g. I also ran the Google sponsored Google Developer Student Club, teaching people about XR, Cloud, and Web technologies for 2 years. I first started becoming really swag once I started attending MIT Reality Hack which I disccovered through running the GDSC, which propelled my career in experimental creative and augmentation technology.
I also once prompted AI to generate a dating website for worms and inspired someone to stageplay my AI-generated youtube dialogues at an off-broadway theater in Brooklyn, so calibrate accordingly.
if... you’re trying to glimpse my polymath range rather quickly*: this account is a mix of XR prototypes, research code, web tools, civic projects, and odd art experiments. The list below is curated, but the repo index is the real proof: github.com/Caerii?tab=repositories.
| project | what | status |
|---|---|---|
| Superintelligent.Group | swarm multiplayer IDE | building |
| MosaicDrone | modular self-assembling drone swarm | designing |
| Assemblies / NEMO | computational cognition (assembly calculus) | researching |
| CommonQuant | market agent societies | testing |
| Unionize.Software | worker coordination tools | organizing |
If you only open a few things, open the repos whose notes tell you how to run it / what to click / what artifact to look at first.
| repo | what | notes |
|---|---|---|
| superintelligent.group | swarm-native collaborative software development environment | open docs/architecture/CONCEPTUAL_OVERVIEW.md for the “VAC” model |
| git2.org | swarm-native version control built on git | run git2-mcp; entity-level locking + intent-aware commits |
| arxiv2.org | “paper → living project” bridge (paste arXiv URL/ID) | start with docker compose up (web+api), then hit /api/resolve |
| SwarmIDE-Alpha | collaborative doc + editor where agents work in the same space as you | start pnpm run server + pnpm run dev (Monaco + Yjs/Y-Sweet) |
| AI-ID | user-owned context layer for AI apps (auth + scoped memory grants) | open ai-id-app/; scopes + store/retrieve APIs; optional pgvector |
| Unionize.Software | privacy-first worker decision-routing + encrypted organizer intake | open the guides corpus via the shipped CLI + local MCP server |
| repo | what | notes |
|---|---|---|
| circuit-tracer | find/visualize/intervene on circuits using transcoder features | open demos/circuit_tracing_tutorial.ipynb (end-to-end) |
| CollaborationCircuitsMechInterp | social cognition / ToM circuits in LLMs (multi-agent settings) | open EXPERIMENTAL_PLAN.md + METHODOLOGY.md first |
| assemblies | neural assembly calculus runtime + research workspace (neural-assemblies) |
run uv run python examples/01_basic_assembly_calculus.py |
| TYPHON | hierarchical memory research harness for long-context / cross-episode eval | run uv run typhon evaluate-memory-suite ... (docs have runbooks) |
| SocietyofScientists | multi-agent grant proposal generator (AG2 + AI21 Jamba + Exa) | start backend, then frontend/ → wizard; export PDF/DOCX/LaTeX |
| feb_2_26_multi_agent_research | local 407-paper multi-agent research corpus + tools | run python search_papers.py \"keyword\" |
| repo | what | notes |
|---|---|---|
| OpenGalea | colocated multiplayer MR controlled by EEG (OpenBCI Cyton + Quest 3) | open Devpost/demo links in the README; hardware + setup sections are real |
| codrawer-bridge | Remarkable Paper Pro strokes → AI ghost-layer ink (co-drawing infra) | start at docs/protocol.md + docs/latency_budget.md |
| DelightfulOS | “internet of bodies” OS: sensors → signal bus → policies → AR overlays | run demo script, then open the dashboard (/dashboard) |
| OpenEgoLens | hackable egocentric camera gateway (stream + control + AI snapshots) | run gateway + UI; frame sampling (1–2s) is the core design choice |
| LifeInBetweenXR | MIT Reality Hack 2023 3D map viewer | open and run the Unity project |
| MicrocosmXR | multiplayer XR sandbox for shared world-building | explore Unity project + Minecraft integration experiments |
| WearableHolographicTheatreKid | wearable holographic theater piece | open the repo and skim the artifacts |
| repo | what | notes |
|---|---|---|
| OpenCAD | modular CAD stack: kernel + solver + feature-tree + agent + viewport | start viewport in mock mode, then flip VITE_USE_MOCK=false for live services |
| SimCorrect | autonomous sim-model fault detection + correction loop | run a Problem*_*/demo.py to watch detect → fix → verify |
| MosaicDrone | “programmable aerial voxels” — modular self-assembling omnidirectional drone swarm | start at “MOSAIC defined” + architecture diagrams in the README |
| OpenEnvironment | natural language → procedural 3D terrain (heightmaps/splatmaps/voxels) | start backend + web, then try the example prompts (“create a desert…”) |
| repo | what | notes |
|---|---|---|
| VisuaML | visual ML editor on the web | open the live site, then the repo for implementation |
| CommonQuant.com | AI-powered market analysis web platform | start the web app and click through the core flows |
| 3d-audio-visualizer | 3D acoustic wave visualization sandbox | start pnpm dev; open src/visualization/IsometricWaveScene.tsx |
| slither.in | dating site for worms + tiny arcade game | open the live site; “Squiggle” is the hidden proof-of-craft |
| smartsight-backend | backend for SmartSight accessibility/vision project | skim routes/services; it’s real backend glue work |
| TheBabelBook | Library of Babel, but with GPT-3 | open the README + core generation logic |
| criticaltheory-wiki | applied critical theory wiki | open the live wiki, then the repo for content mechanics |
| fast360compression | saliency-based 360° video compression research code | run/skim the core C pipeline and results |
This is the “everything else” map — classes, prototypes, websites, one-offs, and experiments that still say something true about how I build.
-
websites & products
- alifjakir.com: my personal experimental blog where I go more into detail on things!
- nanotech.school: interactive web learning for nanotech (work in progress).
- theaicodecompany.com: currently a concept for a portable virtual AI company.
- Caerii.github.io: older GitHub Pages site / web presence experiments.
- CommonQuant.com: the CommonQuant marketing site/app codebase (Next.js).
-
multi-agent / LLM experiments
- AgentMultiverse: experiments with OpenAI Assistants-era agent wiring.
- LanguageAgentTreeSearch: notes/implementation around agentic planning via tree search.
- pdf-parser-ai: “parse PDFs intelligently” prototype.
- PDFScraper: brute-force PDF scraping when you just need the text out.
- AutoTasker: automatic data annotation / task automation experiments.
-
robotics-ish / systems
- DroneDesign: early sketches toward flying-robot design thinking.
- CS444-Final-DiningPhilsophers: concurrency fundamentals (dining philosophers).
- CS444-HW1: parallel thread multiplication (threads + performance basics).
- CS444-HW2: producer/consumer with mutexes (synchronization practice).
- CS444-HW3: keylogger (OS-level input capture exploration).
-
graphics / rendering / play
- -CS452-ComputerGraphicsClass: graphics class work (a trail of shaders and geometry).
- TheMysteryRoom3D: a small 3D world / interactive space experiment.
- GraphicsGame: JavaScript graphics class final (absorb-everything game).
-
audio / media
- 3d-audio-visualizer: a sandbox for 3D audio waveform visualization.
- checkmysoundcloud: music-and-vibes portfolio experiments.
- IS426_Audio2Spectrogram: audio → spectrogram preprocessing pipeline.
-
hackathons / prototypes
- LifeInBetweenXR: MIT Reality Hack 2023 3D map viewer sprint.
- EmergencyAIHelper: HIPAA-ish on-prem emergency AI agents prototype for first responders.
- Zaqathon-submission: hackathon submission repo (prototype snapshot).
-
small utilities / coursework
- XMLParser: simple XML parser (parser-building practice).
- Typescript-Calculator: does what it says on the tin.
- cs460-backend-webapp: backend webapp coursework repo.
- CodePathAndroidPrework: early Android/Kotlin learning app.
-
weird artifacts
- EnviroVoiceImpersonator-EVI: “stick it on objects and it’ll talk” local model physical object "hat" prototype energy.
- slither.in: the worm dating website (yes, really).
- AI_Dialogues_2022: AI-generated dialogues staged as theater.
This is a machine-generated index of my owned repos (sorted by most recent pushes). It’s intentionally exhaustive; the sections above are the human-curated reading path.
open the full repo index (auto-generated)
| repo | what | lang |
|---|---|---|
| slither.in | a dating website for worms | JavaScript |
| CommonQuant.com | AI quantitative trading for everyone | TypeScript |
| Caerii | All about Alif's funky wunky stuff | |
| codrawer-bridge | A low-latency, stroke-native “co-drawer” bridge: reMarkable Paper Pro streams pen strokes to a desktop server; desktop triggers an AI... | Python |
| TYPHON | hierarchical cross-episodic associative memory | Python |
| 3d-audio-visualizer | standalone 3d audio wave visualization sandbox | TypeScript |
| alifjakir.com | My Personal Site | |
| CollaborationCircuitsMechInterp | Mechanistic Interpretability of Multi-Agent Collaboration! | Python |
| OpenGalea | An Open-Source Mixed Reality Brain Computer Interface! Won 5 Quest's from Meta from MIT Reality Hack 2025 for this! | |
| VisuaML | Visual Machine Learning editor on the web bridging category theory, compilers, scientific approaches to AutoML | TypeScript |
| MosaicDrone | A modular, self-assembling omnidirectional drone swarm | Python |
| CS444-Final-DiningPhilsophers | C++ Program that solves the dining philosopher's problem. | C++ |
| Task-Matrix-Flow | Repository for https://replit.com/@AlJ3/Task-Matrix-Flow | TypeScript |
| feb_2_26_multi_agent_research | research tranch | Python |
| MicrocosmXR | MicrocosmXR is a multiplayer XR sandbox with semantic scene understanding for being gods of a virtual civilization. | C# |
| OpenAlterEgo-v0.2 | test draft of open alter ego, for silent speech interfacing | Python |
| LaborRightsNYC | Reforming labor in NYC - evidence and data | TypeScript |
| LifeInBetweenXR | Reality Hack 2023 3D Map Viewer | C# |
| smartsight-backend | spinoff smartsight | JavaScript |
| smartsight-ios | spinoff smartsight | |
| 3D-Web-Sandbox | Physics sandbox in the web! | Makefile |
| AI-ID | Personalized qualitative user context layer for AI apps | TypeScript |
| criticaltheory-wiki | a wiki for the new era of applied critical theory | MDX |
| theaicodecompany.com | concept for a portable virtual AI company | TypeScript |
| unlinkedin.org | work clicker. | TypeScript |
| OpenEnvironment | natural language driven game environment generation system | Python |
| SwarmIDE-Alpha | flow-state enhancing multi-agent project creation | TypeScript |
| Caerii.github.io | HTML | |
| nanotech.school | intuitive ways to learn nanotechnology on the web | TypeScript |
| sylashorowitz | Personal portfolio website | JavaScript |
| pdf-parser-ai | test module to parse PDFs intelligently | TypeScript |
| Kernel-Optimization-Puzzles | Kernel Optimization Challenges | Python |
| Carbon-Fee-and-Dividend-Simulator | carbon fee simulator for changing mass state level policy | Python |
| Zaqathon-submission | TypeScript | |
| High-Frequency-Financial-Modeling-Python | For our own learning and experimentation on financial information | Python |
| Massive-Multiplayer-Vibe-Coding | A novel paradigm for multiplayer vibe coding | |
| StigmergySimulator | cool shader toys | |
| Thinkstruct-Neural-Query | Private Vector DB | TypeScript |
| aaausbangla.com | US Bengali Tourism website | TypeScript |
| Time-Analyzer | Life analytics dashboard stuff | Python |
| StudentSites | My student's websites | TypeScript |
| Team16-Hackathon.Bio | Automatic Data Scientist Society of Agents | |
| Artificialgeneralintelligence-dev-website | AGI Dev community | JavaScript |
| Identicon-Generator-Site | Creating identicons automatically! | TypeScript |
| Locus | This creates a novel process for accelerating utility infrastructure siting through real-time, comprehensive data access, communities, an... | |
| fast360compression | Georgia State University - REU 2021 Saliency Based Compression Technique for 360 Videos | C |
| GrimRepo | Alif and Eloise killed the repository | |
| RecipeCalculator | Help plan recipes and meals over weeks to make sure you get all nutrients in! | Python |
| Clarkson-Google-Dev-Site | Maintaining the Website for it's evolution into the MIT GDSC | JavaScript |
| WhyHowBeta | Experimental Implementation of WhyHow.AI Graph Toolkits for AI Understanding | Python |
| InvestComply | prototype | Python |
| EmergencyAIHelper | HIPPA Compliant On-Premise Emergency AI Agents Prototype Implementation for First Responders to create effort reduction for those in real... | |
| WearableHolographicTheatreKid | Epic wearable theatre kid made by Alif Jakir & Sylas Horowitz | JavaScript |
| suno-api | TypeScript | |
| Scratchpad | Just some testing stuff | |
| EmberAlifGame-RenameLater | Experimental Implementations for an interesting game-thing! Add more desc. later! | ReScript |
| AutoTasker | Automatically Does Data Annotation Tasks | |
| checkmysoundcloud | Music and Vibes portfolio | JavaScript |
| Obscure-Trial | Hypothetical Gumroad Design | CSS |
| DroneDesign | We explore new frontiers in flying robots! | |
| AgentMultiverse | Alif Jakir and Ryan Tran's experiments with OpenAI Assistants | Python |
| PDFScraper | It scrapes PDFs, what more do you want? | Python |
| AlexAlifThoughts | A repository of thoughts and directions for future research | |
| LanguageAgentTreeSearch | LANGUAGE AGENT TREE SEARCH UNIFIES REASONING ACTING AND PLANNING IN LANGUAGE MODELS | |
| EnviroVoiceImpersonator-EVI | Stick it on objects and stuff! It’ll talk… or else | |
| -CS452-ComputerGraphicsClass | Some cool stuff that I did in class! | JavaScript |
| TheBabelBook | The Library of Babel, now with GPT-3! | Python |
| Low-Light-Enhancement-GAN | DTU Computational Photography Project | Python |
| Photonics_Processing | For July 2023 Denmark Technical University Machine Vision Course | |
| Typescript-Calculator | does what it says on the tin | |
| CS473-ComputerVisionClass | Here are ways that we made computers see! | Python |
| cs460-backend-webapp | JavaScript | |
| CS444-HW3 | Keylogger | C++ |
| XMLParser | Simple Parser for XML documents | Python |
| AI_Dialogues_2022 | Performed As a Play on 2/26/23 in Brooklyn, NYC | |
| CS444-HW2 | Producer Consumer Problem with mutexes | C++ |
| CS444-HW1 | Parallel Thread Multiplication | C++ |
| CodePathAndroidPrework | Simple silly android application for learning purposes | Kotlin |
| DolbyVideoCallDemoTestXR | TypeScript | |
| Logseq | Up to date Logseq for my second digital brain | |
| IS426_Audio2Spectrogram | Processes audio into usable spectrograms for classification | Python |
| TheMysteryRoom3D | This room is unlike any you've ever seen before. | |
| GraphicsGame | A game for our CS452 Graphics Class. You are the president of the school and you need to absorb everyone and everything. | JavaScript |
| ClackPublic | Allows for Client-Server Communication with a GUI | HTML |
| LogseqOld | All my networked notes! | |
| GPT3WebSummarizer | Create easy summaries of web content. | C# |
| logseq-1 | A privacy-first, open-source platform for knowledge sharing and management. | Clojure |
| GitHelloDate | Testing Github | Java |
open the forks index (top recent)
Forks of things I am exploring, working on contributions to, or are saving (sorted by recent pushes; top 60 shown).
| repo | what | lang |
|---|---|---|
| agents-are-thinking | A weekend project exploring agent "thinking" animation. | Python |
| WaveletLM | WaveletLM | Python |
| vortexnet | VortexNet: Neural Computing through Fluid Dynamics | Python |
| ast-outline | Fast, AST-based structural outline for source files. Built for LLM coding agents and humans. | Rust |
| LogseqVisionClaw | Logseq Spring Thing Immersive & Agentic Knowledge Development Engine | Rust |
| TileKernels | A kernel library written in tilelang | Python |
| fleetmdm | Open device management (live: https://fleetdm.com) | Go |
| spikingjelly | SpikingJelly is an open-source deep learning framework for Spiking Neural Network (SNN) based on PyTorch. (live: https://spikingjelly.readthedocs.io) | Python |
| minebench | Minecraft-style voxel benchmark for comparing AI models (Arena + Sandbox) (live: https://minebench.ai) | TypeScript |
| late-sh | A cozy terminal clubhouse for developers. Lofi beats, casual games, chat, and tech news, all via SSH. (live: https://late.sh) | Rust |
| catgrad | a categorical deep learning compiler | Rust |
| ZWM | Zero-shot World Models Are Developmentally Efficient Learners | Python |
| obsidian-wiki | Framework for AI agents to build and maintain an Obsidian wiki using Karpathy's LLM Wiki pattern | Python |
| codex | Lightweight coding agent that runs in your terminal | Rust |
| claude-mem | A Claude Code plugin that automatically captures everything Claude does during your coding sessions, compresses it with AI (using Claude'... (live: https://claude-mem.ai) | TypeScript |
| og-logseq | Logseq og (file version) (live: https://logseq.com) | Clojure |
| open-agents | An open source template for building cloud agents. (live: https://open-agents.dev) | TypeScript |
| b3d | Bayes3D | Jupyter Notebook |
| bayes3d | (live: https://probcomp.github.io/bayes3d/) | Jupyter Notebook |
| stretchystudio | FOSS 2D animation tool for turning static illustrations into mesh-deformable characters (live: https://stretchy.studio) | JavaScript |
| -Project-Nord-Spiking-Neural-Network-Language-Model | The first pure SNN language model trained from scratch with a fully original architecture. 618M parameters • 93% sparsity • Runs on p... (live: https://www.nord-ai.net/) | Python |
| rattles | 🪇 Minimal terminal spinners for Rust (live: https://docs.rs/rattles) | Rust |
| promnesia | Another piece of your extended mind (live: https://beepb00p.xyz/promnesia.html) | Python |
| neuroskill | NeuroSkill™ — State of Mind Brain-Computer Interface system (live: https://neuroskill.com) | Rust |
| playcanvas-engine | Powerful web graphics runtime built on WebGL, WebGPU, WebXR and glTF (live: https://playcanvas.com) | JavaScript |
| zigzag | fast source code documentation (live: https://zagforge.com/) | Zig |
| graphiti | Build Real-Time Knowledge Graphs for AI Agents (live: https://help.getzep.com/graphiti) | Python |
| xrtracker-package | AR object tracking plugin for Unity. Supports AR Foundation (iOS/Android) and desktop webcams. Silhouette, depth, and edge-based tracking... | C# |
| zep-python | Build Agents That Recall What Matters. Systematically engineer relevant context from chat history & business data. (Python Client) (live: https://help.getzep.com) | Python |
| playcanvas-web-sdk | Gracia Web SDK | HTML |
| tribev2 | This repository contains the code to train and evaluate TRIBE v2, a multimodal model for brain response prediction | Jupyter Notebook |
| pretext | TypeScript | |
| titans-trainer | Train TITANS models in 5 lines. Huggingface-style trainer for the TITANS architecture - transformers with neural long-term memory & test-... | Python |
| eips | The efficient, intention-preserving sequence CRDT. Guaranteed logarithmic-time operations. | Rust |
| git4aiagents | All reasoning is stored in git. No external server. No database. No account. No hosting. Any developer on any system pulls the latest co... (live: https://www.git4aiagents.com/) | Python |
| ironclaw | IronClaw is OpenClaw inspired implementation in Rust focused on privacy and security (live: https://www.ironclaw.com) | Rust |
| dioxus | Fullstack app framework for web, desktop, and mobile. (live: https://dioxuslabs.com) | Rust |
| mmr | Python based algorithmic trading platform for Interactive Brokers | Python |
| hyperspace-db | [H] HyperspaceDB is a high-performance, hyperbolic vector database written in Rust. It features 1-bit quantization, async replication, an... | Rust |
| ngrep | grep plus word embeddings | Rust |
| LabClaw | LabClaw – Operating Layer for LabOS (Stanford-Princeton AI Co-Scientists) (live: https://labclaw-ai.github.io/) | |
| three-geospatial | Geospatial Rendering in Three.js (live: https://takram-design-engineering.github.io/three-geospatial) | TypeScript |
| stable-worldmodel | Reliable, minimal and scalable library for evaluating and conducting world model research (live: https://galilai-group.github.io/stable-worldmodel/) | Python |
| slowrun | 100M tokens. Infinite compute. Lowest val loss wins. | Python |
| scienceclaw | Python | |
| scg_Foundation-1-comfyUI | ComfyUI nodes for RoyalCities Foundation-1 — structured text-to-sample music generation with full BPM, key, timbre, and FX control | Python |
| autoresearch-go-ane | Autonomous AI research on Apple Silicon using ANE-accelerated training | Go |
| superml | An ML engineering plugin for your coding agents. (live: https://leeroo.com/) | Python |
| Infinite | TypeScript | |
| meetily | Privacy first, AI meeting assistant with 4x faster Parakeet/Whisper live transcription, speaker diarization, and Ollama summarization bui... (live: https://meetily.ai) | Rust |
| ml_forge | A visual-based graph node editor for training computer vision models. | Python |
| RaTeX | KaTeX-compatible LaTeX math renderer in pure Rust. No JavaScript, no WebView, no DOM. One Rust core → iOS, Android, Flutter, ... (live: https://erweixin.github.io/RaTeX/demo/support_table.html) | Rust |
| causationentropy | Implementation of Causation Entropy from Clarkson Center for Complex Systems Science (C3S2) (live: https://www.clarkson.edu/academics/research/labs-centers/c3s2) | Python |
| rustdesk | An open-source remote desktop application designed for self-hosting, as an alternative to TeamViewer. (live: https://rustdesk.com) | Rust |
| Awesome-DLMs | The official GitHub repo for the survey paper "A Survey on Diffusion Language Models". (live: https://arxiv.org/abs/2508.10875) |



