├── LICENSE └── README.md /LICENSE: -------------------------------------------------------------------------------- 1 | Apache License 2 | Version 2.0, January 2004 3 | http://www.apache.org/licenses/ 4 | 5 | TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION 6 | 7 | 1. Definitions. 8 | 9 | "License" shall mean the terms and conditions for use, reproduction, 10 | and distribution as defined by Sections 1 through 9 of this document. 11 | 12 | "Licensor" shall mean the copyright owner or entity authorized by 13 | the copyright owner that is granting the License. 14 | 15 | "Legal Entity" shall mean the union of the acting entity and all 16 | other entities that control, are controlled by, or are under common 17 | control with that entity. For the purposes of this definition, 18 | "control" means (i) the power, direct or indirect, to cause the 19 | direction or management of such entity, whether by contract or 20 | otherwise, or (ii) ownership of fifty percent (50%) or more of the 21 | outstanding shares, or (iii) beneficial ownership of such entity. 22 | 23 | "You" (or "Your") shall mean an individual or Legal Entity 24 | exercising permissions granted by this License. 25 | 26 | "Source" form shall mean the preferred form for making modifications, 27 | including but not limited to software source code, documentation 28 | source, and configuration files. 29 | 30 | "Object" form shall mean any form resulting from mechanical 31 | transformation or translation of a Source form, including but 32 | not limited to compiled object code, generated documentation, 33 | and conversions to other media types. 34 | 35 | "Work" shall mean the work of authorship, whether in Source or 36 | Object form, made available under the License, as indicated by a 37 | copyright notice that is included in or attached to the work 38 | (an example is provided in the Appendix below). 39 | 40 | "Derivative Works" shall mean any work, whether in Source or Object 41 | form, that is based on (or derived from) the Work and for which the 42 | editorial revisions, annotations, elaborations, or other modifications 43 | represent, as a whole, an original work of authorship. For the purposes 44 | of this License, Derivative Works shall not include works that remain 45 | separable from, or merely link (or bind by name) to the interfaces of, 46 | the Work and Derivative Works thereof. 47 | 48 | "Contribution" shall mean any work of authorship, including 49 | the original version of the Work and any modifications or additions 50 | to that Work or Derivative Works thereof, that is intentionally 51 | submitted to Licensor for inclusion in the Work by the copyright owner 52 | or by an individual or Legal Entity authorized to submit on behalf of 53 | the copyright owner. For the purposes of this definition, "submitted" 54 | means any form of electronic, verbal, or written communication sent 55 | to the Licensor or its representatives, including but not limited to 56 | communication on electronic mailing lists, source code control systems, 57 | and issue tracking systems that are managed by, or on behalf of, the 58 | Licensor for the purpose of discussing and improving the Work, but 59 | excluding communication that is conspicuously marked or otherwise 60 | designated in writing by the copyright owner as "Not a Contribution." 61 | 62 | "Contributor" shall mean Licensor and any individual or Legal Entity 63 | on behalf of whom a Contribution has been received by Licensor and 64 | subsequently incorporated within the Work. 65 | 66 | 2. Grant of Copyright License. Subject to the terms and conditions of 67 | this License, each Contributor hereby grants to You a perpetual, 68 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 69 | copyright license to reproduce, prepare Derivative Works of, 70 | publicly display, publicly perform, sublicense, and distribute the 71 | Work and such Derivative Works in Source or Object form. 72 | 73 | 3. Grant of Patent License. Subject to the terms and conditions of 74 | this License, each Contributor hereby grants to You a perpetual, 75 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 76 | (except as stated in this section) patent license to make, have made, 77 | use, offer to sell, sell, import, and otherwise transfer the Work, 78 | where such license applies only to those patent claims licensable 79 | by such Contributor that are necessarily infringed by their 80 | Contribution(s) alone or by combination of their Contribution(s) 81 | with the Work to which such Contribution(s) was submitted. If You 82 | institute patent litigation against any entity (including a 83 | cross-claim or counterclaim in a lawsuit) alleging that the Work 84 | or a Contribution incorporated within the Work constitutes direct 85 | or contributory patent infringement, then any patent licenses 86 | granted to You under this License for that Work shall terminate 87 | as of the date such litigation is filed. 88 | 89 | 4. Redistribution. You may reproduce and distribute copies of the 90 | Work or Derivative Works thereof in any medium, with or without 91 | modifications, and in Source or Object form, provided that You 92 | meet the following conditions: 93 | 94 | (a) You must give any other recipients of the Work or 95 | Derivative Works a copy of this License; and 96 | 97 | (b) You must cause any modified files to carry prominent notices 98 | stating that You changed the files; and 99 | 100 | (c) You must retain, in the Source form of any Derivative Works 101 | that You distribute, all copyright, patent, trademark, and 102 | attribution notices from the Source form of the Work, 103 | excluding those notices that do not pertain to any part of 104 | the Derivative Works; and 105 | 106 | (d) If the Work includes a "NOTICE" text file as part of its 107 | distribution, then any Derivative Works that You distribute must 108 | include a readable copy of the attribution notices contained 109 | within such NOTICE file, excluding those notices that do not 110 | pertain to any part of the Derivative Works, in at least one 111 | of the following places: within a NOTICE text file distributed 112 | as part of the Derivative Works; within the Source form or 113 | documentation, if provided along with the Derivative Works; or, 114 | within a display generated by the Derivative Works, if and 115 | wherever such third-party notices normally appear. The contents 116 | of the NOTICE file are for informational purposes only and 117 | do not modify the License. You may add Your own attribution 118 | notices within Derivative Works that You distribute, alongside 119 | or as an addendum to the NOTICE text from the Work, provided 120 | that such additional attribution notices cannot be construed 121 | as modifying the License. 122 | 123 | You may add Your own copyright statement to Your modifications and 124 | may provide additional or different license terms and conditions 125 | for use, reproduction, or distribution of Your modifications, or 126 | for any such Derivative Works as a whole, provided Your use, 127 | reproduction, and distribution of the Work otherwise complies with 128 | the conditions stated in this License. 129 | 130 | 5. Submission of Contributions. Unless You explicitly state otherwise, 131 | any Contribution intentionally submitted for inclusion in the Work 132 | by You to the Licensor shall be under the terms and conditions of 133 | this License, without any additional terms or conditions. 134 | Notwithstanding the above, nothing herein shall supersede or modify 135 | the terms of any separate license agreement you may have executed 136 | with Licensor regarding such Contributions. 137 | 138 | 6. Trademarks. This License does not grant permission to use the trade 139 | names, trademarks, service marks, or product names of the Licensor, 140 | except as required for reasonable and customary use in describing the 141 | origin of the Work and reproducing the content of the NOTICE file. 142 | 143 | 7. Disclaimer of Warranty. Unless required by applicable law or 144 | agreed to in writing, Licensor provides the Work (and each 145 | Contributor provides its Contributions) on an "AS IS" BASIS, 146 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or 147 | implied, including, without limitation, any warranties or conditions 148 | of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A 149 | PARTICULAR PURPOSE. You are solely responsible for determining the 150 | appropriateness of using or redistributing the Work and assume any 151 | risks associated with Your exercise of permissions under this License. 152 | 153 | 8. Limitation of Liability. In no event and under no legal theory, 154 | whether in tort (including negligence), contract, or otherwise, 155 | unless required by applicable law (such as deliberate and grossly 156 | negligent acts) or agreed to in writing, shall any Contributor be 157 | liable to You for damages, including any direct, indirect, special, 158 | incidental, or consequential damages of any character arising as a 159 | result of this License or out of the use or inability to use the 160 | Work (including but not limited to damages for loss of goodwill, 161 | work stoppage, computer failure or malfunction, or any and all 162 | other commercial damages or losses), even if such Contributor 163 | has been advised of the possibility of such damages. 164 | 165 | 9. Accepting Warranty or Additional Liability. While redistributing 166 | the Work or Derivative Works thereof, You may choose to offer, 167 | and charge a fee for, acceptance of support, warranty, indemnity, 168 | or other liability obligations and/or rights consistent with this 169 | License. However, in accepting such obligations, You may act only 170 | on Your own behalf and on Your sole responsibility, not on behalf 171 | of any other Contributor, and only if You agree to indemnify, 172 | defend, and hold each Contributor harmless for any liability 173 | incurred by, or claims asserted against, such Contributor by reason 174 | of your accepting any such warranty or additional liability. 175 | 176 | END OF TERMS AND CONDITIONS 177 | 178 | APPENDIX: How to apply the Apache License to your work. 179 | 180 | To apply the Apache License to your work, attach the following 181 | boilerplate notice, with the fields enclosed by brackets "[]" 182 | replaced with your own identifying information. (Don't include 183 | the brackets!) The text should be enclosed in the appropriate 184 | comment syntax for the file format. We also recommend that a 185 | file or class name and description of purpose be included on the 186 | same "printed page" as the copyright notice for easier 187 | identification within third-party archives. 188 | 189 | Copyright [yyyy] [name of copyright owner] 190 | 191 | Licensed under the Apache License, Version 2.0 (the "License"); 192 | you may not use this file except in compliance with the License. 193 | You may obtain a copy of the License at 194 | 195 | http://www.apache.org/licenses/LICENSE-2.0 196 | 197 | Unless required by applicable law or agreed to in writing, software 198 | distributed under the License is distributed on an "AS IS" BASIS, 199 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 200 | See the License for the specific language governing permissions and 201 | limitations under the License. 202 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # awesome-ai-engineering 2 | List of resources helping you become a better AI engineer. 3 | 4 | # What's an AI Engineer? 5 | 6 | * [Microsoft's definition](https://learn.microsoft.com/en-us/training/career-paths/ai-engineer) "Artificial intelligence (AI) engineers are responsible for developing, programming and training the complex networks of algorithms that make up AI so that they can function like a human brain. This role requires combined expertise in software development, programming, data science and data engineering" 7 | * [Coursera's definition](https://www.coursera.org/articles/ai-engineer) "Artificial intelligence engineers are individuals who use AI and machine learning techniques to develop applications and systems that can help organizations increase efficiency, cut costs, increase profits, and make better business decisions." 8 | * [Tech Target](https://www.techtarget.com/whatis/feature/How-to-become-an-artificial-intelligence-engineer) "AI engineers develop, program and train the complex networks of algorithms that encompass AI so those algorithms can work like a human brain. AI engineers must be experts in software development, data science, data engineering and programming." 9 | * [Swyx podcast](https://pca.st/episode/ada36079-7bfa-4f06-9020-cdf37e65e34f?t=498.0) (17 April 2024) 10 | * [Scaler Blogs](https://www.scaler.com/blog/how-to-become-an-ai-engineer/) "AI engineers design, develop, and deploy intelligent systems using machine learning, deep learning, and NLP to solve complex problems and enable autonomous decision-making." 11 | 12 | # What's the difference between an AI Engineer and a Machine Learning Engineer? 13 | 14 | * [UpWork](https://www.upwork.com/resources/ai-engineer-vs-ml-engineer#:~:text=AI%20engineers%20work%20on%20a,predictions%20from%20large%20data%20sets.) "AI engineers work on a broader set of tasks that encompass various forms of machine intelligence, like neural networks, to develop AI models for specific applications. In contrast, ML engineers focus more on ML algorithms and models that can self-tune to better learn and make predictions from large data sets." 15 | 16 | # What's the difference between an AI Engineer and a Software Engineer? 17 | * [IEEE](https://www.computer.org/csdl/magazine/so/2022/06/09928183/1HJux2cbygM) ChatGPT's summary of that page "**AI engineers blend traditional software engineering skills with a deep understanding of machine learning and artificial intelligence** to develop systems that enhance decision-making and automation within organizations. They are proficient in AI technologies and statistical analysis, focusing on building and integrating AI models into applications. On the other hand, software engineers focus broadly on designing, implementing, and maintaining software systems, with a comprehensive grasp of the software development lifecycle, from requirement analysis to deployment and maintenance. The distinction is further marked by the AI engineer's need to navigate emerging AI technologies, whereas software engineers adhere to established engineering principles and practices across various platforms and technologies" 18 | 19 | # Practical Tools & Techniques 20 | 21 | This section covers useful stuff you can use to become a better AI engineer. 22 | 23 | ## LLM Platforms and APIs 24 | ### LLM Platforms 25 | - ChatGPT 26 | - Claude.ai 27 | - Phind (dev focus, GPT4+own) 28 | - Microsoft Copilot (GPT4+own) 29 | - Perplexity.ai 30 | - You.com 31 | - groq.com 32 | 33 | ### LLM APIs and Inference Services 34 | 35 | #### GPU Marketplaces 36 | - [GPUList.ai](https://gpulist.ai/) 37 | - [Vast.ai](https://vast.ai) 38 | -[Prime Intellect](https://www.primeintellect.ai/) 39 | 40 | 41 | #### free open weight playgrounds 42 | Try out open source models instantly. 43 | - [Perplexity Labs](https://labs.perplexity.ai) side by side comparison 44 | - [Groq chat](https://chat.groq.com) demo a subset of models on Groq's proprietary inference hardware (LPUs) 45 | - [Vercel AI Playground](https://sdk.vercel.ai/) 46 | 47 | #### self-hosted Open weight inference 48 | - ollama (go/open source) 49 | - [LocalAI](https://github.com/mudler/LocalAI) (go/open source) 50 | - msty.app 51 | - Nitro.jan.ai 52 | - [Paddler](https://github.com/distantmagic/paddler) scaling / load balancing of llama.cpp inference 53 | 54 | #### SaaS 55 | - [fal.ai](https://fal.ai) 56 | - [lepton.ai](https://www.lepton.ai/) 57 | - modal.com: on demand Serverless container +GPU execution runtime 58 | - Predibase: LLM fine-tuning and hosting 59 | - [https://hpc-ai.com/](HPC AI): GPU rental 60 | - Replicate.com: models-as-a service 61 | - Together.ai: Serverless LLM / multimodal inference 62 | - Lambda Labs: Manual rental of GPUs / clusters 63 | - Beam.cloud: Serverless generative AI fast standup 64 | - Runpod 65 | - [Cloudflare Workers AI](https://blog.cloudflare.com/workers-ai) 66 | - Coreweave: autoscale GPU + Serverless (knative) 67 | - Mosaicml: (acquired by Databricks) 68 | - mixedbread.ai: retrieval as a service (search, reranking, embedding) 69 | - lamini.ai: LLM inference 70 | - Anyscale + rai.ai scaling 71 | - HF inference API 72 | - massedcompute.com 73 | - Salad.com 74 | - Openpipe.ai 75 | - Unsloth.ai 76 | - Crusoe.ai GPU rental 77 | - Akash 78 | - Groq: ultra fast LLM for selected models 79 | - BoltAI 80 | - [Saturn Cloud](https://saturncloud.io/) 81 | - Fireworks.ai 82 | - Inferless.com 83 | - Banana.dev (defunct) 84 | - pipeline.ai 85 | - [hyperstack.cloud](https://www.hyperstack.cloud/) 86 | - [Alibaba Elastic GPU service](https://www.alibabacloud.com/en/product/heterogeneous_computing?_p_lc=1) 87 | - [Cloudalize GPU Kubenetes Service](https://www.cloudalize.com/solutions/kubernetes-gpu-cloud/) 88 | - [Tensordock.com](https://tensordock.com/benchmarks) 89 | - [Fly GPU](https://fly.io/gpu) GPUs on demand 90 | - [Jarvis Labs](https://jarvislabs.ai/) GPUs on demand 91 | - [BentoML](https://bentoml.com/) open source open weight inference with cloud option 92 | - [bitbop GPU dev the cloud](https://bitbop.io/) 93 | - [Simplepid.ai](SimplePod) 94 | 95 | ### Structured output 96 | - SGLang 97 | - outlines 98 | - Instructor 99 | - Marginalia 100 | 101 | ### Prompt engineering 102 | - [ATLAS](https://github.com/VILA-Lab/ATLAS) 103 | - DSPy 104 | - [Microsoft llmlingua prompt compression](https://github.com/microsoft/LLMLingua) 105 | 106 | 107 | ## LLM Development and Optimization 108 | ### LLM Testing and Evaluation 109 | - promptfoo 110 | - Ollama grid search 111 | - Uptrain 112 | - Google Cloud GCP AutoSxS 113 | - Paloma 114 | - LightEval 115 | - Bayesian Evaluation 116 | - Mozilla's experience 117 | - Ruler (long context evaluation) 118 | - OpenAI Simple Evals 119 | - [Moonshot](https://github.com/aiverify-foundation/moonshot) 120 | 121 | 122 | 123 | #### Leaderboards / Evaluations 124 | - [SEAL](https://scale.com/leaderboard) 125 | - Lmsys.org 126 | - [HuggingFace Open LLM leaderboard](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) 127 | - [Vectara Hallucination Leaderboard](https://huggingface.co/spaces/vectara/leaderboard) 128 | - [Text Embedding Leaderboard](https://huggingface.co/spaces/mteb/leaderboard) 129 | - [Enterprise Use Case Leaderboard](https://huggingface.co/spaces/PatronusAI/enterprise_scenarios_leaderboard) Finance, Legal, Customer Support 130 | - [MixEval](https://mixeval.github.io/) 131 | - [Arena Hard Auto](https://github.com/lm-sys/arena-hard-auto) 132 | - [Google Instruction Following Eval / IFEval](https://github.com/google-research/google-research/tree/master/instruction_following_eval) 133 | 134 | #### Observability 135 | - [Phoenix](https://github.com/Arize-ai/phoenix) 136 | - [Helicone](https://www.helicone.ai/) 137 | 138 | ### Pretraining 139 | [llm.c: Andrey Karparthy's GPT-2 from thr ground up in raw C](https://x.com/yuchenj_uw/status/1798594515168903307?s=46) 140 | 141 | ### Human Input Methods 142 | - RLHF 143 | - DPO 144 | - TKO 145 | - LIPO 146 | - DORA 147 | - SPO 148 | 149 | ### Architecture Innovations 150 | - Longformer 151 | - Reformer 152 | - BigBird 153 | - Attention Beacons 154 | - RWKV 155 | - Denseformer 156 | - [Microsoft SliceGPT](https://github.com/microsoft/TransformerCompression) remove up to 25% of layers 157 | - [DCFormer](https://github.com/Caiyun-AI/DCFormer) 158 | 159 | #### Tokenizers 160 | - [ZeTT](https://github.com/bminixhofer/zett) 161 | 162 | ### Fine-Tuning and Optimization 163 | - Lazy Axolotl 164 | - Lit-GPT 165 | - Predibase 166 | - Fine Tune Llama 2 Colab (by HF) 167 | - Openpipe.ai 168 | - LISA 169 | - Torchtune 170 | - LASER layer reduction 171 | - lmstudio.ai 172 | - [AutoQuant](https://colab.research.google.com/drive/1b6nqC7UZVt8bx4MksX7s656GXPM-eWw4?usp=sharing) 173 | - [Mistral fine tuning service](https://mistral.ai/news/customization/) [Github](https://github.com/mistralai/mistral-finetune) 174 | 175 | ### Task-Optimized LLMs and Context Extension 176 | - Predibase LORALand 177 | - RoPE 178 | - Ailibi 179 | - LongRoPE 180 | - Unsloth+RoPE 181 | - [InfiniAttention](https://github.com/kyegomez/Infini-attention): a pathway to ultra long context windows with manageable memory consumption 182 | 183 | ## Infrastructure and Tools 184 | ### Vector Stores / Information Retrieval 185 | - pinecone 186 | - weaviate 187 | - chroma (open source) 188 | - lancedb (open source) 189 | - postgresql + pgvector (open source) 190 | - sqlite + vss (open source) 191 | - faiss by meta 192 | - Vespa.ai + binary embeddings 193 | 194 | #### Telemetry 195 | - [IR measures](https://github.com/terrierteam/ir_measures) 196 | 197 | ### Cloud Hosting 198 | - Blueocean / paperspace for GPU 199 | - AWS 200 | - GCP 201 | - Azure 202 | - Hetzner GPU 203 | - Cloudflare 204 | 205 | ### Notebooks and Code Interpreters 206 | - Lightning Studio 207 | - Google Colab 208 | - ChatGPT 209 | - Julius.ai 210 | 211 | ### Attention Mechanisms 212 | - FlashAttentionv2 213 | - HippoAttention 214 | - RingAttention 215 | - PagedAttention 216 | 217 | ### Model Merging 218 | - Efficient Linear Model Merging for LLMs 219 | - Automerge 220 | - Sakana Evolutionary Model Merge 221 | 222 | ### Optimizers and Autodifferentiation 223 | #### Optimizers 224 | - Adam 225 | - AdamW 226 | - Prodigy 227 | - Schedule-free optimizers (April 2024) 228 | 229 | #### Autodifferentiation Libraries 230 | - SymPy 231 | - torch.autograd 232 | - Autograd 233 | - tf.GradientTape 234 | - gomlx 235 | 236 | ### Prompt Debugging 237 | - mitmproxy (via Show Me The Prompt) 238 | 239 | ### Agents and Swarms 240 | - CrewAI 241 | - Autogen 242 | - OpenDevin 243 | - [SWE-agent](https://github.com/princeton-nlp/SWE-agent) 244 | - [Leda](https://github.com/elder-plinius/Leda) 245 | - [Devon](https://github.com/entropy-research/Devon) open source pair programmer 246 | - [HuggingFace Agents](https://huggingface.co/docs/transformers/main/en/agents) 247 | 248 | ### Analytics 249 | - [Agent Ops](https://github.com/AgentOps-AI/agentops) 250 | - [Weights and Biases](https://wandb.com) 251 | - [Okareo](https://okareo.com/) 252 | 253 | ### Chat with Your Data/RAG 254 | - Weaviate [Verba](https://github.com/weaviate/Verba): RAG solution using Weaviate 255 | - Microsoft GitHub 256 | - AWS Bedrock embeddings, streamlit, langchain, pinecone, claude, etc. 257 | - AWS Serverless 258 | - GCP 259 | - Gemini for document processing 260 | - AWS knowledge bases for bedrock 261 | - [FLARE](https://github.com/jzbjyb/FLARE) dynamically replace low-probability tokens with RAG lookups 262 | - [Embedchain](https://github.com/embedchain/embedchain) 263 | 264 | ### Guardrails and Safety 265 | 266 | #### Protection 267 | - Llamaguard 268 | - Llamaguard with streaming 269 | - Guardrails for AWS Bedrock 270 | 271 | #### Jailbreaks 272 | - [Pliny the Prompter jailbreaks](https://github.com/elder-plinius/L1B3RT45) 273 | - [Jailbreak LLMs](https://github.com/verazuo/jailbreak_llms) 274 | - Haize 275 | 276 | ### Embeddings and Document Processing 277 | #### Embeddings Services 278 | - Amazon Titan Embeddings 279 | - Huggingface 280 | - Nomic + ollama 281 | - Cohere multi-aspect embeddings 282 | - LLM2Vec 283 | 284 | #### Document Extraction Services 285 | - Amazon Kendra 286 | 287 | #### Embeddings Algorithms 288 | - Colbert 289 | - Binary quantization (BitNet) 290 | 291 | ### Multi-Adapter Models 292 | For hosting multiple fine-tunes at once 293 | - Punica 294 | 295 | ### GPU Usage Optimization 296 | - Run.ai -- service for bare metal GPU cluster management now owned by Nvidia 297 | 298 | ### Important Datasets 299 | - sst2 sentiment movie sentiment (HF) 300 | - 650,000 English books 301 | - Openwebtext 302 | - Fineweb 303 | 304 | ### Synthetic Data Generation 305 | - generator9000 306 | 307 | ### GPUs and Accelerators 308 | - Groq 309 | - Truffle-1 310 | 311 | ### Data Curation 312 | - NeMo-Curator 313 | 314 | ### ML Local Mini Clusters 315 | - Tinybox / tinygrad 316 | - WOPR (7 x 4090) 317 | 318 | ### Data Labeling 319 | - Argilla Distilabel 320 | - Spacy Prodigy 321 | - Snorkel 322 | - [Refuel-AI autolabel](https://github.com/refuel-ai/autolabel) 323 | 324 | ### Model Configuration Management 325 | - DVCorg 326 | - WandB Weave 327 | --------------------------------------------------------------------------------