BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Huntsville AI - ECPv6.8.3//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Huntsville AI
X-ORIGINAL-URL:https://hsv.ai
X-WR-CALDESC:Events for Huntsville AI
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:20240310T080000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:20241103T070000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:20250309T080000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:20251102T070000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250301T000000
DTEND;TZID=America/Chicago:20250301T235959
DTSTAMP:20260405T184846
CREATED:20241216T013024Z
LAST-MODIFIED:20250222T160525Z
UID:1820-1740787200-1740873599@hsv.ai
SUMMARY:HudsonAlpha Tech Challenge
DESCRIPTION:
URL:https://hsv.ai/event/hudsonalpha-tech-challenge-2/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/12/hatch-1.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250219T180000
DTEND;TZID=America/Chicago:20250219T190000
DTSTAMP:20260405T184846
CREATED:20241216T012737Z
LAST-MODIFIED:20250219T211911Z
UID:1818-1739988000-1739991600@hsv.ai
SUMMARY:VIRTUAL ONLY - Deep Research Overview
DESCRIPTION:This week we will have Jack Francis provide an overview of Deep Research: \nWe will explore Google’s Gemini Deep Research\, OpenAI’s Deep Research\, and Open-Source Deep Research implementations. Deep Research is an LLM agent that uses reasoning to synthesize large amounts of online information and complete multi-step research tasks for you. We’ll dive into the mechanics of Deep Research\, discussing how these agents function and some of the benchmarks used to evaluate the models. \nWe will discuss the strengths and limitations of each approach\, provide examples for comparison\, and talk through Deep Research use cases\, such as market research! Deep Research is useful tool to add to your toolbox for market research\, decision analysis\, or any other task that requires significant amounts of “Googling” today. The goal of the presentation is to introduce Deep Research and provide an overview of current tools\, so that you can use it effectively for your own research tasks. \nPaper Review Series \nThanks to Josh for the great kickoff to the paper review series last week! You can find the recording\, links to papers\, and transcript here: \nhttps://hsv.ai/2025/02/12/ai-reasoning-reinforcement-learning-fundamental-papers-history/ \nIf you have a paper that you would like to see us review\, please let me know. We’ll post the date for March as soon as it’s finalized.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks & Other Events: \n\n2025 HATCH – https://hudsonalpha.org/techchallenge/\n2025 BSides – https://nac-issa.org/events/bsides/\n\nDetails: \n\nDate – 2/19/2025\nTime – 6-7:30pm\nLocation – VIRTUAL\nZoom –https://us02web.zoom.us/j/84476860693?pwd=D2UGBsfpb5ZKoJEGMiFZaroV9aEb28.1
URL:https://hsv.ai/event/deep-research-overview/
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/12/Deep-Research-Overview.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250212T180000
DTEND;TZID=America/Chicago:20250212T193000
DTSTAMP:20260405T184846
CREATED:20250209T145141Z
LAST-MODIFIED:20250209T145141Z
UID:1851-1739383200-1739388600@hsv.ai
SUMMARY:Virtual Paper Review - Reasoning
DESCRIPTION:Join us virtually this Wednesday at 6pm to kick off our monthly Paper Review series! We will be holding these sessions every 4 weeks on Zoom. The first one will be led by Josh Phillips: \nThis month we will be exploring the current evolution in AI reasoning models. We will start with a short review of the history of combining reinforcement learning with tree search for reasoning (AlphaGo/MuZero). Then we will cover the real start of the reasoning race with the chain-of-thought innovations highlighted by “Let’s Verify Step by Step.” And finally of course\, we’ll cover the most important insights from DeepSeek’s three recent papers that build on all of this prior foundation: \n\nDeepSeek-V3 – a highly efficient MoE-based approach (671B parameters\, only 37B active per token) powered by advanced CUDA/NCCL optimizations and low-precision (FP8) training\, making huge models tractable on constrained hardware.\nDeepSeek-R1 – the first open demonstration that an LLM can learn complex reasoning purely through RL rewards\, without any annotated examples (r1-zero)\nDeepSeekMath– which introduces GRPO (an alternative to PPO that simplifies training and removes the need for a separate value network).\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks: \n\nMastering Atari\, Go\, Chess and Shogi by Planning with a Learned Model (2019) :  https://arxiv.org/pdf/1911.08265\nLets Verify Step by Step (2023) : https://arxiv.org/abs/2305.20050\nDeepseek V3 Technical Report (2024): https://github.com/deepseek-ai/DeepSeek-V3/blob/main/DeepSeek_V3.pdf\nDeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning  (2025) https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf\nDeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models (2025): https://arxiv.org/pdf/2402.03300\n(Bonus) “We’ve got R1 at home” : https://unsloth.ai/blog/r1-reasoning\n\nDetails: \n\nDate – 2/12/2025\nTime – 6-7:30pm\nLocation – VIRTUAL\nZoom –https://us02web.zoom.us/j/86964741971?pwd=SXmv4xfY8M6TZbEOIXyYfLu5Y3FAil.1
URL:https://hsv.ai/event/virtual-paper-review-reasoning/
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2025/02/Paper-Series-Reasoning.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250205T180000
DTEND;TZID=America/Chicago:20250205T190000
DTSTAMP:20260405T184846
CREATED:20241216T012527Z
LAST-MODIFIED:20250203T052119Z
UID:1816-1738778400-1738782000@hsv.ai
SUMMARY:DeepSeek-R1 and Other Topics
DESCRIPTION:This week we will be back at HudsonAlpha in their classroom. This is a new room for us\, which I believe is on the opposite side from the large conference room we have been using. \nHudsonAlpha Tech Challenge \nWe should have Tyler Clark stop by for a few minutes to discuss some changes to HATCH and how we can get involved. Otherwise\, I can cover based on what I know. The main change from the past challenges is that this will be a one day challenge occurring on Saturday\, March 1st. \nDeepSeek-R1 \nWe’ll be discussing DeepSeek-R1\, which has been unavoidable in most conversations over the last week – and rightly so. We will cover the architecture\, the approach used for training\, as well as the distilled models based on the Llama and Qwen architectures. We have several folks in the group that already have DeepSeek-R1 running locally\, so if you need help getting it running or would like to connect directly please let me know. \nFor part of the discussion\, you may want to refer back to the Mixture of Experts talk that Josh Phillips led during one of our meetups in October 2023. You can find the link below. \nPaper Review Series \nThis year we are going to start a review series to look at published papers in-depth. These will be a monthly\, virtual-only meetup to start with but may shift based on interest. The first paper review will be on Wednesday\, February 12\, led by Josh Phillips to cover the DeepSeek-R1 paper (also linked below). \nSBIR/STTR Topics? \nThe next round of SBIR/STTR Topics are supposed to drop on Wednesday. Given the issues with funding being stopped for NSF grants and other changes driven by the new administration\, I don’t know if we will see new topics drop or not. \nAI Symposium \nWe’ll take a few minutes to talk about how the AI Symposium sessions went.  The organizers are interested in feedback for the quality of the sessions and any suggested changes for next year.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks & Other Events: \n\nDeepSeek-R1 on GitHub – https://github.com/deepseek-ai/DeepSeek-R1?tab=readme-ov-file\nDeepSeek-R1 Paper – https://arxiv.org/pdf/2501.12948\nMixture of Experts Discussion – https://hsv.ai/2023/10/25/mixture-of-experts-harnessing-the-hidden-architecture-of-gpt4/\nIntroduction to AI Slides – https://github.com/HSV-AI/presentations/blob/master/2025/250116_Introduction_to_AI_LearningQuest.pdf\nUses of AI Slides – https://github.com/HSV-AI/presentations/blob/master/2025/250123_Uses_of_AI_LearningQuest.pdf?raw=true\n2025 HATCH – https://hudsonalpha.org/techchallenge/\n2025 BSides – https://nac-issa.org/events/bsides/\n\nDetails: \n\nDate – 2/05/2025\nTime – 6-7:30pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/82283427713?pwd=PCuFpoeIIYQGhAWDain4YK4FoRbETU.1
URL:https://hsv.ai/event/deepseek-r1/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/12/DeepSeek-R1.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250130T130000
DTEND;TZID=America/Chicago:20250130T143000
DTSTAMP:20260405T184846
CREATED:20241216T011949Z
LAST-MODIFIED:20241216T011949Z
UID:1810-1738242000-1738247400@hsv.ai
SUMMARY:Intro to AI @ LearningQuest
DESCRIPTION:
URL:https://hsv.ai/event/intro-to-ai-learningquest-3/
LOCATION:Huntsville Madison County Library\, 915 Monroe St\, Huntsville\, AL\, 35801\, United States
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/12/LearningQuest.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250128T000000
DTEND;TZID=America/Chicago:20250130T235959
DTSTAMP:20260405T184846
CREATED:20241216T012353Z
LAST-MODIFIED:20241216T012353Z
UID:1812-1738022400-1738281599@hsv.ai
SUMMARY:2025 AI Symposium: Automation Vs HITL
DESCRIPTION:
URL:https://hsv.ai/event/2025-ai-symposium-automation-vs-hitl/
LOCATION:Space & Rocket Center\, 1 Tranquility Base\, Huntsville\, AL\, 35805\, United States
ATTACH;FMTTYPE=image/jpeg:https://hsv.ai/wp-content/uploads/2024/12/SRC-AI-Symposium.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250123T130000
DTEND;TZID=America/Chicago:20250123T143000
DTSTAMP:20260405T184846
CREATED:20241216T011814Z
LAST-MODIFIED:20241216T011814Z
UID:1808-1737637200-1737642600@hsv.ai
SUMMARY:Intro to AI @ LearningQuest
DESCRIPTION:
URL:https://hsv.ai/event/intro-to-ai-learningquest-2/
LOCATION:Huntsville Madison County Library\, 915 Monroe St\, Huntsville\, AL\, 35801\, United States
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/12/LearningQuest.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250122T180000
DTEND;TZID=America/Chicago:20250122T190000
DTSTAMP:20260405T184846
CREATED:20241216T010345Z
LAST-MODIFIED:20250120T042451Z
UID:1800-1737568800-1737572400@hsv.ai
SUMMARY:AI for Code Development
DESCRIPTION:It has certainly been a busy few weeks! The first session for LearningQuest (slides linked below) had about 150 people in attendance and went very well. I had several great questions come up that I will have to research before the next session on Thursday. \nWe don’t know exactly what the weather will do this week as far as closures. We have the big conference room at HudsonAlpha reserved\, but please check for further updates before coming in person in case we switch to a virtual only meetup for this week. \nThe main subject of the night will be AI for code development. We looked at this some time in the past with the StarCoder model\, but there are many options available now. Some are free (and limited)\, while others are inexpensive. We will try to cover Copilot\, Cursor\, and a self hosted option. Please let me know if you have experience/opinions on these and we can get that in the discussion as well! \nThere is also an NVIDIA meetup at Torch earlier on Wednesday. I believe that several of us might be able to make it\, so we should be able to share any info we pick up. I’ll drop a link to their signup in case you want to go. \nLinks & Other Events: \n\nIntroduction to AI Slides – https://github.com/HSV-AI/presentations/blob/master/2025/250116_Introduction_to_AI_LearningQuest.pdf\nNVIDIA Meetup at Torch – https://torchtechnologies.surveymonkey.com/r/CSHQHBQ\n2025 AI Symposium – https://www.rocketcenter.com/institute \n2025 HATCH – https://hudsonalpha.org/techchallenge/\n2025 BSides – https://nac-issa.org/events/bsides/\n\nDetails: \n\nDate – 1/22/2025\nTime – 6-7:30pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/89635091670?pwd=FFO9xzQrA3cl2fXxGYWXuPenLQIUcU.1
URL:https://hsv.ai/event/ai-for-code-dev/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/12/AI-for-Code-Development.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250116T130000
DTEND;TZID=America/Chicago:20250116T143000
DTSTAMP:20260405T184846
CREATED:20241216T011717Z
LAST-MODIFIED:20241216T011831Z
UID:1804-1737032400-1737037800@hsv.ai
SUMMARY:Intro to AI @ LearningQuest
DESCRIPTION:
URL:https://hsv.ai/event/intro-to-ai-learningquest/
LOCATION:Huntsville Madison County Library\, 915 Monroe St\, Huntsville\, AL\, 35801\, United States
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/12/LearningQuest.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250108T180000
DTEND;TZID=America/Chicago:20250108T190000
DTSTAMP:20260405T184846
CREATED:20241216T010036Z
LAST-MODIFIED:20241216T011337Z
UID:1798-1736359200-1736362800@hsv.ai
SUMMARY:2025 Kickoff
DESCRIPTION:
URL:https://hsv.ai/event/2025-kickoff/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/12/2025-Kickoff.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20241218T180000
DTEND;TZID=America/Chicago:20241218T190000
DTSTAMP:20260405T184846
CREATED:20241030T031438Z
LAST-MODIFIED:20241216T153222Z
UID:1774-1734544800-1734548400@hsv.ai
SUMMARY:2024 Year in Review
DESCRIPTION:We made it through 2024! Well\, close enough for this group anyway 🙂 This week we will get together and go through the list of topics and other items that we covered over the year. There are 3 questions that I want to ask for each topic: \n\nDid it stick\, or was it overcome by other technology or approaches?\nWhat is the “shelf life” of the topic?\nWhat direction is the topic headed? Advancing\, decay\, community interest?\n\nIn addition to the review of 2024\, we also have Paul Bay coming to talk about what he has been doing with AI Agents. Based on what I am reading\, Agentic AI could be the next popular movement in the field. \nWe’ll end the session by talking about plans for 2025. The kickoff is scheduled for January 8th. I have placeholders for other meetups & associated events on the calendar through February if you want to take a look here: https://hsv.ai/calendar/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks & Other Events: \n\n2024 AI Symposium Recorded Sessions – https://www.youtube.com/playlist?list=PLvvHQqQynqmtkQuLsvfFtmy0OohBcA5H4\n2025 AI Symposium – https://www.rocketcenter.com/institute \n2025 HATCH – https://hudsonalpha.org/techchallenge/\n\n\nDetails: \n\nDate – 12/18/2024\nTime – 6-7:30pm\nZoom –https://us02web.zoom.us/j/83751396321?pwd=ZdQYvlGBEn2vcr5Bmb9XjrPrQhJhAa.1
URL:https://hsv.ai/event/2024-year-in-review/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/10/2024-Review-Agents.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20241204T180000
DTEND;TZID=America/Chicago:20241204T190000
DTSTAMP:20260405T184846
CREATED:20241030T031139Z
LAST-MODIFIED:20241203T155515Z
UID:1772-1733335200-1733338800@hsv.ai
SUMMARY:AI Application Scaffolding
DESCRIPTION:This week we will walk through the components used in the transcription service along with how they are integrated and deployed. This has been updated over the year to incorporate new features and infrastructure as it became available. \nThis session will be a little light on AI\, but should give you an idea of what is involved in making a product that is consumable by the general public. There is likely too much to cover in an hour\, so we’ll start with the basic framework and dive in wherever folks want. Also – if you join us for the meetup\, I’ll share the code behind all of this. \nBasic Technologies: \n\nDocker – separate containers for UI and transcription service\nFaster Whisper – does the transcription\nParagraph Detection – finds the best breaks\nSummarization – does it hallucinate\, you bet!\nFlask – web server behind the UI\nReact – Typescript based UI\nSocketIO – live updates between web server and UI\nPytest – making sure stuff works\nTerraform – couldn’t do AWS without this\nStripe – how to get paid\n\nAmazon Web Service Components: \n\nAWS Container Registry\nAWS CloudWatch\nAWS Lambda\nAWS S3\nAWS Fargate\nAWS SQS\nAWS SES\nAWS DynamoDB\nAWS AIM\n\nAlso – the recordings for the last two sessions are now posted at https://hsv.ai/videos/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks & Other Events: \n\n2024 AI Symposium Recorded Sessions – https://www.youtube.com/playlist?list=PLvvHQqQynqmtkQuLsvfFtmy0OohBcA5H4\n2025 AI Symposium – https://www.rocketcenter.com/institute \n\nDetails: \n\nDate – 12/04/2024\nTime – 6-7:30pm\nZoom –https://us02web.zoom.us/j/81815701970?pwd=yatSAPMVsU7delXnrvA3N0zk218Gg1.1
URL:https://hsv.ai/event/ai-application-scaffolding/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/10/AI-Application-Scaffolding.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20241120T180000
DTEND;TZID=America/Chicago:20241120T190000
DTSTAMP:20260405T184846
CREATED:20241030T030833Z
LAST-MODIFIED:20241120T050446Z
UID:1769-1732125600-1732129200@hsv.ai
SUMMARY:Overview of Text to Speech Approaches
DESCRIPTION:Thanks to Josh Phillips for hosting last week while I was out! I should have the recording up soon at https://hsv.ai/videos/ \nThis week we will take a look at Speech to Text models in three different categories: \n\nProducts that create audio from text in an offline mode\nAPIs that can be integrated into a product\nOpen source models that you can host\n\nEach of these presents different challenges that we’ll cover such as latency\, realism\, and hallucination. Here’s the list of products and models so far\, so if you don’t see your favorite in the list\, let me know and we’ll check it out as well: \n\nParler\nCoqui\nBark\nOpenAI (6 models)\nBASE TTS (Amazon)\nMetaVoice\nMeloTTS\nElevenLabs\nFacebook MMS\n\nAlso – a few of us went to the Huntsville AI and Machine Learning Technology Exchange and Expo last week\, so we might do an overview of those topics if time permits.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks & Other Events: \n\n2024 AI Symposium Recorded Sessions – https://www.youtube.com/playlist?list=PLvvHQqQynqmtkQuLsvfFtmy0OohBcA5H4\n2025 AI Symposium – https://www.rocketcenter.com/institute\nHugging Face Text to Speech – https://huggingface.co/tasks/text-to-speech \n\nDetails: \n\nDate – 11/20/2024\nTime – 6-7:30pm\nZoom –https://us02web.zoom.us/j/89971705398?pwd=CndWhnWX6sbtgQLaAbn8CctPjzcxjV.1\n\nAs always\, I really appreciate the support and replies to these emails. You can also help by following\, sharing\, liking\, and dropping comments on my posts on LinkedIn and Facebook – especially the ones directly for the Huntsville AI page on LinkedIn – https://www.linkedin.com/company/huntsville-ai
URL:https://hsv.ai/event/text-to-speech/
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/10/Text-to-Speech.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20241106T180000
DTEND;TZID=America/Chicago:20241106T190000
DTSTAMP:20260405T184846
CREATED:20241030T030545Z
LAST-MODIFIED:20241102T174915Z
UID:1766-1730916000-1730919600@hsv.ai
SUMMARY:Structured Outputs with Finite State Machines & Open Source Speech to Speech
DESCRIPTION:We have two guest speakers this week\, Josh Phillips & Charlie Rogers! Josh has covered a few topics for us in the past\, but this will be Charlie’s first time presenting for us\, so please come and engage. \nTopic 1 (Josh Phillips) : Structured Outputs with Finite State Machines \nDiscover how the integration of finite state machines with guided decoding can significantly enhance the stability and reliability of Large Language Models’ inference processes. In this session\, you will delve into the intricacies of structured generation\, learning how it not only ensures strict adherence to desired formats but also accelerates processing speeds by up to five times. \nExperience firsthand demonstrations of various pipelines transitioning from unstructured to structured outputs\, illustrating the ease and effectiveness of implementation. Join us to unlock advanced techniques that optimize both the performance and consistency of your language models. \nTopic 2 (Charlie Rogers) : Let’s Talk It Out : Open Source Speech to Speech (S2S) \nDive into the world of open-source speech-to-speech systems and discover how to build efficient pipelines from scratch. Learn about the modular approach\, explore user-friendly tools\, and learn from Charlie’s experience with automating client-server architectures using RunPod. Ideal for developers and researchers looking to advance their speech processing projects. \n2025 AI Symposium: \nWe should also have representatives from the Space & Rocket Center AI Symposium stop by to give an introduction to the symposium for next year. Last year was a lot of fun\, and this year appears to be even bigger. I’ll drop a link to last years sessions below.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks & Other Events: \n\n2024 AI Symposium Recorded Sessions – https://www.youtube.com/playlist?list=PLvvHQqQynqmtkQuLsvfFtmy0OohBcA5H4\n2025 AI Symposium – https://www.rocketcenter.com/institute\nHuntsville AI and Machine Learning Technology Exchange and Expo – https://meetingsevents.reg.ext.hpe.com/event/0144e3b2-0951-4eea-b9a9-21be49da3441/summary\nAI Innovators of Huntsville (Andrey’s group) at GigaParts –https://events.gigaparts.com/events/gigapartshuntsville/1422131\n\nDetails: \n\nDate – 11/06/2024\nTime – 6-7:30pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/85119272393?pwd=A90PaqdUl1SEz6hyB8PFDbhTN3dKTB.1\n\nWe had a fantastic turnout for the social\, and it was great to meet a lot of new people coming from different backgrounds. Thanks to Phillip Lee with ISSA for boosting our exposure with the local cyber community! \nAs always\, I really appreciate the support and replies to these emails. You can also help by following\, sharing\, liking\, and dropping comments on my posts on LinkedIn and Facebook – especially the ones directly for the Huntsville AI page on LinkedIn – https://www.linkedin.com/company/huntsville-ai \nHope to see you soon!\n-J.
URL:https://hsv.ai/event/structured-output/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/10/2topics-1.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20241023T180000
DTEND;TZID=America/Chicago:20241023T193000
DTSTAMP:20260405T184846
CREATED:20241018T140501Z
LAST-MODIFIED:20241018T140501Z
UID:1760-1729706400-1729711800@hsv.ai
SUMMARY:Fall Social
DESCRIPTION:Our next meetup with be a social at Stovehouse\, where we can get to know each other a bit better. We had a large group last week and it’s difficult to catch up and learn what everyone is doing with AI in such a short period of time. \nIf you’re new to the group\, or haven’t been to a meetup\, this is a great way to get involved and see who is doing fun things with AI. \nFrom last week – the video is posted along with a transcript – see the Links below. It’s always interesting to see what the transcription engine does with terms in a southern accent. This time it’s “BLLM” instead of “vLLM”. \nIn other news – there is a “Huntsville AI and Machine Learning Technology Exchange and Expo” coming up on November 13 at the Westin Hotel. While it’s not affiliated with this group\, it is a really good conference with good speakers. It’s free to go\, and if you would like to register as part of Huntsville AI\, let me know and I can get you the name & address to use. You should be able to register personally for free as well. \nLinks: \n\nvLLM\, SBIR Topics\, Helene Response – https://hsv.ai/2024/10/09/vllm-sbir-topics-helene-response/\nHuntsville AI and Machine Learning Technology Exchange and Expo – https://meetingsevents.reg.ext.hpe.com/event/0144e3b2-0951-4eea-b9a9-21be49da3441/summary\n\n 
URL:https://hsv.ai/event/fall-social/
LOCATION:Stovehouse\, 3414 Governors Dr SW\, Huntsville\, AL\, 35805
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/10/Social.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20241009T180000
DTEND;TZID=America/Chicago:20241009T190000
DTSTAMP:20260405T184846
CREATED:20241007T052406Z
LAST-MODIFIED:20241007T052406Z
UID:1751-1728496800-1728500400@hsv.ai
SUMMARY:vLLM\, New SBIR Topics\, Helene Response
DESCRIPTION:We have a few topics that we will cover this week\, led by Josh Phillips (vLLM) and Gary Turner (Helene Response). We will also cover the list of AI related SBIR topics that were released last week. \nvLLM\nIn this session\, we will demonstrate how easy it is to deploy vLLM on RunPod using an A40 GPU instance (approximately $0.40 per hour for 48GB of VRAM). We will be deploying the new LLama 3.2 11B vision model to showcase vLLM’s deployment simplicity and performance capabilities in a cost-effective environment. This hands-on example will provide practical insights into getting started with vLLM and show the real-world benefits of using a platform like RunPod for efficient model serving. \nHurricane Helene Response\nGary and I have been having a running conversation about the chaotic response in the Eastern TN and Western NC mountains. This will be a short brainstorming session to see if there’s anything we can do to assist. \nSBIR/STTR\nAccording to the announcement on the DoD SBIR/STTR website: “New DoD SBIR & STTR topics will be released on the first Wednesday of every month.” To align with this schedule\, we will stop planning our normal sessions specific to SBIR and cover new topics as they come in. From the list of topics that dropped last week\, the AI related topics are: \n\nAF244-0001: Interactive Knowledge Graphs for Situational Awareness\nAF244-0004: Securing Legacies: Developing an AI-Enhanced Privacy Architecture for Military Family Knowledge Systems\nAF244-D002: Automated Correction of Optical Distortions in Canopy Transparencies\nAF244-D006: Artificial Intelligence-Enabled Design Rule Checker\nAF244-D016: Develop next-generation autonomous tracking\nOSD254-P001: Generative AI BAA Open Topic\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks: \n\nvLLM – https://docs.vllm.ai/en/latest/\nRunPod – https://www.runpod.io/\n\n\nDetails: \n\nDate – 10/09/2024\nTime – 6-7pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/89843812772?pwd=YS6OlaDTMKKkwzuRvd7uk6oLIQ2Qce.1
URL:https://hsv.ai/event/vllm-new-sbir-topics-helene-response/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/10/vLLM.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240925T180000
DTEND;TZID=America/Chicago:20240925T190000
DTSTAMP:20260405T184846
CREATED:20240925T154526Z
LAST-MODIFIED:20240925T154526Z
UID:1744-1727287200-1727290800@hsv.ai
SUMMARY:Pixtral Model by Mistral AI
DESCRIPTION:We will cover the Pixtral multimodal model released by Mistral AI last week. Hopefully we can get it loaded up in Le Chat (French ChatGPT?) and kick the tires on it a bit. From what I am hearing from some of our group\, it’s definitely something to look into. \nFrom the Mistral announcement: \n\nNatively multimodal\, trained with interleaved image and text data\nStrong performance on multimodal tasks\, excels in instruction following\nMaintains state-of-the-art performance on text-only benchmarks\nArchitecture:\n\nNew 400M parameter vision encoder trained from scratch\n12B parameter multimodal decoder based on Mistral Nemo\nSupports variable image sizes and aspect ratios\nSupports multiple images in the long context window of 128k tokens\n\n\nUse:\n\nLicense: Apache 2.0\nTry it on La Plateforme or on Le Chat\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks: \n\nPixtral – https://mistral.ai/news/pixtral-12b/\nLe Chat – https://mistral.ai/news/le-chat-mistral/\nHuggingFace – https://huggingface.co/mistralai/Pixtral-12B-2409\n\n\nDetails: \n\nDate – 09/25/2024\nTime – 6-7pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/81091066071?pwd=U166DpG5fuaObEjeHb1tmi0oz8s9g2.1
URL:https://hsv.ai/event/pixtral-model-by-mistral-ai/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/09/Pixtral.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240911T180000
DTEND;TZID=America/Chicago:20240911T190000
DTSTAMP:20260405T184846
CREATED:20240909T040928Z
LAST-MODIFIED:20240909T040928Z
UID:1732-1726077600-1726081200@hsv.ai
SUMMARY:Faster Whisper for Audio Transcription
DESCRIPTION:This week we will cover the Faster Whisper library. It is similar to the approach used by llama.cpp and provides a much smaller footprint and faster execution times than the OpenAI Whisper library. \nWe will also walk through the approach for using this library within an AWS Lambda function\, along with tips and tricks for deploying in a docker container. \nAlso – I’ll be joining the Hsv.py meetup on September 19 to talk about Llama-cpp-python if you would like to join us next week as well. More information in the links below.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks: \n\nFaster Whisper – https://github.com/SYSTRAN/faster-whisper\nAWS Lambda Limitations – https://blog.awsfundamentals.com/lambda-limitations\nHSV.py meetup – https://www.meetup.com/hsv-py/events/302429808/\n\n\nDetails: \n\nDate – 09/11/2024\nTime – 6-7pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/86564069248?pwd=nEig8eeg2tuTcjQsMQIWaZCJ1yZvnF.1
URL:https://hsv.ai/event/faster-whisper-for-audio-transcription/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/09/Faster-Whisper.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240821T180000
DTEND;TZID=America/Chicago:20240821T190000
DTSTAMP:20260405T184846
CREATED:20240129T035424Z
LAST-MODIFIED:20240818T022331Z
UID:1557-1724263200-1724266800@hsv.ai
SUMMARY:2024 SBIR Topics Round 3
DESCRIPTION:This week we will cover the last round of DoD SBIR Topics that are scheduled to be released on Wednesday. This is always a good checkpoint to see what areas of AI that DoD organizations are interested in. \nThe last time we covered SBIR topics\, there were several on the call that were new to this approach of funding research. If you are also in this boat – here’s a good resource from SBIR.gov that covers the basics: https://www.sbir.gov/about \nYou can think of it as seed funding\, with the initial round being somewhere around $100k for a 6 month effort\, followed by a potential $2M project for 2 years. Each agency and topic may have a different funding allotment\, so be sure and check the details of any topic that you are interested in. \nOur normal approach is to use the topic search provided here – https://www.dodsbirsttr.mil/topics-app/ but this time we will also see what happens if we ask a LLM to analyze the topics for us.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks: \n\nSBIR Topics and Topic Search – https://www.dodsbirsttr.mil/topics-app/\nPrevious SBIR Awards – https://www.sbir.gov/awards\n\n\nDetails: \n\nDate – 08/21/2024\nTime – 6-7pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/86939079628?pwd=yb6JZ1IZJCjouX6BYfGWf1a9o90azS.1\n\n\n\n\n\n\n\n\n\n  \nPRE-RELEASE: August 21\, 2024\nOPEN: September 18\, 2024\nCLOSE: October 16\, 2024
URL:https://hsv.ai/event/2024-sbir-topics-round-3/
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2023/01/SBIR-Topics.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240807T180000
DTEND;TZID=America/Chicago:20240807T190000
DTSTAMP:20260405T184846
CREATED:20240505T034806Z
LAST-MODIFIED:20240505T034806Z
UID:1659-1723053600-1723057200@hsv.ai
SUMMARY:SMD Symposium Roundup
DESCRIPTION:
URL:https://hsv.ai/event/smd-symposium-roundup/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240724T180000
DTEND;TZID=America/Chicago:20240724T190000
DTSTAMP:20260405T184846
CREATED:20240505T034709Z
LAST-MODIFIED:20240722T035343Z
UID:1657-1721844000-1721847600@hsv.ai
SUMMARY:Private RAG Deployment & Cost
DESCRIPTION:We have been covering a series this year based on how a RAG works\, and how to build one. Now it’s time to look at what it will cost to deploy it! \nBLUF: If you have constraints that require you to host your own LLM instead of using services such as OpenAI\, Anthropic\, or Google – this is going to get expensive. \nThe baseline that we will be using to derive requirements and prices will be the a subset documents that we have been using along the way from NASA. To start\, I have pulled 500 of these documents which total 109\,516 paragraphs and 4\,254\,032 words. \nWe will start with a basic use case with using OpenAI GPT4o and a Weaviate hosted vector store. We will move from there to a self hosted solution for all of the components and add up the month cost associated.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSeries: \nOver the next several sessions\, we will be diving deeper into separate components needed for RAG – hopefully resulting in a chat-based Q&A service for the NASA Technical Report Server.  We were introduced to this data during our submission for the 2022 NASA SpaceApps Challenge – where we placed 2nd. Our submission was a semantic search based on the abstracts for the NSTR dataset of 10\,000 papers. \nLinks: \n\nHuntsville AI 2022 SpaceApps Submission – https://github.com/HSV-AI/spaceapps2022\n\n\nDetails: \n\nDate – 07/24/2024\nTime – 6-7pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/84452278503?pwd=rJwxSbD1EAUdIHuzGoMscHYxpfULhR.1
URL:https://hsv.ai/event/private-rag-deployment-cost-2/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/05/Private-RAG.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240710T180000
DTEND;TZID=America/Chicago:20240710T190000
DTSTAMP:20260405T184846
CREATED:20240505T034618Z
LAST-MODIFIED:20240708T112827Z
UID:1655-1720634400-1720638000@hsv.ai
SUMMARY:RAG Prompt Engineering
DESCRIPTION:If you have ever tried to prompt an LLM to give you what you want\, you have probably run into some frustrating challenges. This frustration can range from trying to write what you want in pseudo-code to arguing with a child about putting their toys away. \nThis week we will focus on some guidelines for building successful prompts for a RAG system. There is a lot of art and craft associated\, and the answer to most questions is “it depends”. To help understand this better\, we will pull back the curtain and talk about WHY certain prompts are effective and what types of situations they may apply to. \nI am learning more about this subject every day\, so if you have prior experience or tips that I should include – please reply and I’ll add them to the talk.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSeries: \nOver the next several sessions\, we will be diving deeper into separate components needed for RAG – hopefully resulting in a chat-based Q&A service for the NASA Technical Report Server.  We were introduced to this data during our submission for the 2022 NASA SpaceApps Challenge – where we placed 2nd. Our submission was a semantic search based on the abstracts for the NSTR dataset of 10\,000 papers. \nLinks: \n\nHuntsville AI 2022 SpaceApps Submission – https://github.com/HSV-AI/spaceapps2022\nOpenAI Discussion – https://community.openai.com/t/prompt-engineering-for-rag\nPrompt Engineering Guide – https://www.promptingguide.ai/\nOpenAI Cookbook – https://cookbook.openai.com/\n\n\nDetails: \n\nDate – 07/10/2024\nTime – 6-7pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/82515370839?pwd=66LA4Gpn9WXecEZzK4aS6Mnaghj1tC.1
URL:https://hsv.ai/event/rag-prompt-engineering/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/05/RAG-Prompts.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240626T180000
DTEND;TZID=America/Chicago:20240626T190000
DTSTAMP:20260405T184846
CREATED:20240505T034359Z
LAST-MODIFIED:20240624T034406Z
UID:1653-1719424800-1719428400@hsv.ai
SUMMARY:Vector Storage with Weaviate
DESCRIPTION:We’re going to add in a new topic to the NASA RAG series this week\, so check below for some rearranging of the schedule. As always\, if there is something that you would like to talk about (or have someone cover) – please let me know! \nAfter a few rounds of iteration using ChromaDB for vector storage\, it has become apparent that there are better options. While it is possible to run ChromaDB locally\, it has too large of a footprint to run alongside the other components to develop a RAG prototype. I’ve spent a few hours replacing the ChromaDB integration with Weaviate\, and I’m quite happy with the results. \nIn this session\, we’ll cover everything you need to know to get started with Weaviate. There are several options that I’m not using for the NASA RAG\, but may be relevant for other types of products that you may want to build.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSeries: \nOver the next several sessions\, we will be diving deeper into separate components needed for RAG – hopefully resulting in a chat-based Q&A service for the NASA Technical Report Server.  We were introduced to this data during our submission for the 2022 NASA SpaceApps Challenge – where we placed 2nd. Our submission was a semantic search based on the abstracts for the NSTR dataset of 10\,000 papers. \nLinks: \n\nHuntsville AI 2022 SpaceApps Submission – https://github.com/HSV-AI/spaceapps2022\nhttps://weaviate.io/platform\nhttps://github.com/weaviate/weaviate\n\nDetails: \n\nDate – 06/26/2024\nTime – 6-7pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/89078627184?pwd=x2RxNAGKubipwXmjc1tYaPUf3gKcHD.1
URL:https://hsv.ai/event/vector-storage-weaviate/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/05/Weaviate.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240612T180000
DTEND;TZID=America/Chicago:20240612T190000
DTSTAMP:20260405T184846
CREATED:20240505T034251Z
LAST-MODIFIED:20240608T180816Z
UID:1651-1718215200-1718218800@hsv.ai
SUMMARY:Law and AI
DESCRIPTION:This week we have Andrew Tuggle from Womble Bond Dickinson (US) LLP joining us to talk about Law and AI. He has been involved in our meetups for years and always has an interesting perspective from an attorney’s point of view. \nAs AI becomes more advanced and ubiquitous\, it is raising new legal issues across many domains. In this session\, we will explore key legal considerations surrounding AI through the following topics: \n\nPersonhood and agency of AI systems\nIntellectual property issues including authorship\, derivative works\, and open source\nAI decisioning and regulatory oversight\nRegulatory schemes – Biden executive order and the AI Act in the European Union\n\nAlso – we will resume the RAG series again following this week:\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSeries: \nOver the next several sessions\, we will be diving deeper into separate components needed for RAG – hopefully resulting in a chat-based Q&A service for the NASA Technical Report Server.  We were introduced to this data during our submission for the 2022 NASA SpaceApps Challenge – where we placed 2nd. Our submission was a semantic search based on the abstracts for the NSTR dataset of 10\,000 papers. \nLinks: \n\nHuntsville AI 2022 SpaceApps Submission – https://github.com/HSV-AI/spaceapps2022\n\n\nDetails: \n\nDate – 06/12/2024\nTime – 6-7pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/84283975695?pwd=cmlm2MzYB95tvRt3M1SAc4b56FE3oy.1\n\nHope to see you soon!\n-J.
URL:https://hsv.ai/event/law-and-ai/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/05/law-and-ai.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240529T180000
DTEND;TZID=America/Chicago:20240529T200000
DTSTAMP:20260405T184846
CREATED:20240505T034153Z
LAST-MODIFIED:20240528T022456Z
UID:1649-1717005600-1717012800@hsv.ai
SUMMARY:Summer Social
DESCRIPTION:We are going to take a break from the formal sessions this week to get together at Stovehouse and hang out for a bit. This is a great opportunity to relax and meet others involved in Huntsville AI. We have a wide variety of backgrounds\, so I always learn something interesting. This officially starts at 6pm\, but I will be there early if anyone wants to help hold a table. \nWe’re suspending the RAG series for the next two sessions\, but will resume at the end of June. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSeries: \nOver the next several sessions\, we will be diving deeper into separate components needed for RAG – hopefully resulting in a chat-based Q&A service for the NASA Technical Report Server.  We were introduced to this data during our submission for the 2022 NASA SpaceApps Challenge – where we placed 2nd. Our submission was a semantic search based on the abstracts for the NSTR dataset of 10\,000 papers. \nLinks: \n\nhttps://www.stovehouse.com/\nHuntsville AI 2022 SpaceApps Submission – https://github.com/HSV-AI/spaceapps2022\n\n\nDetails: \n\nDate – 05/29/2024\nTime – 6-8pm\nLocation – Stovehouse – most likely near the Pourhouse
URL:https://hsv.ai/event/summer-social/
LOCATION:Stovehouse\, 3414 Governors Dr SW\, Huntsville\, AL\, 35805
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2022/05/220316-Social.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240515T180000
DTEND;TZID=America/Chicago:20240515T190000
DTSTAMP:20260405T184846
CREATED:20240505T033848Z
LAST-MODIFIED:20240513T035034Z
UID:1644-1715796000-1715799600@hsv.ai
SUMMARY:OpenAI API
DESCRIPTION:This week we will cover the use of the OpenAI API with both llama-cpp-python and OpenAI itself. After this session\, you will have a solid understanding of how to use the chat completion API along with an approach to run the same code locally against llama-cpp-python. \nWe will cover account setup\, billing & budgets\, as well as the python code needed to make function calls. Next we will show the same python code running against a locally hosted model. \nUpdates: \n\nThe videos page at https://hsv.ai/videos/ is updated through our last meetup\, so if you want to catch up on previous sessions – give that a shot.\nThe meetups through the summer months are all planned – check the schedule toward the end of this email\, as well as https://hsv.ai/calendar/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSeries: \nOver the next several sessions\, we will be diving deeper into separate components needed for RAG – hopefully resulting in a chat-based Q&A service for the NASA Technical Report Server.  We were introduced to this data during our submission for the 2022 NASA SpaceApps Challenge – where we placed 2nd. Our submission was a semantic search based on the abstracts for the NSTR dataset of 10\,000 papers. \nLinks: \n\nhttps://platform.openai.com/docs/api-reference\nhttps://github.com/abetlen/llama-cpp-python\nhttps://llama-cpp-python.readthedocs.io/en/latest/\nHuntsville AI 2022 SpaceApps Submission – https://github.com/HSV-AI/spaceapps2022\n\n\nDetails: \n\nDate – 05/15/2024\nTime – 6-7pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/81984208192?pwd=VkYySnpjQ0hOblB2bEt3c1JlMXdBdz09
URL:https://hsv.ai/event/openai-api/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240501T180000
DTEND;TZID=America/Chicago:20240501T190000
DTSTAMP:20260405T184846
CREATED:20240428T135334Z
LAST-MODIFIED:20240428T135334Z
UID:1640-1714586400-1714590000@hsv.ai
SUMMARY:Deep Dive with Llama-Cpp-Python
DESCRIPTION:This week we will take another look at llama-ccp\, this time using the python wrapper available in llama-cpp-python. This library can be installed and imported directly\, or run as a separate service available through the OpenAI API. \nWe covered the basics of llama-cpp last year and looked through the different quantization approaches available. It provides a great way to run a lower level model using only CPU resources and RAM. There is some quality loss due to the quantization\, but the results are good enough for prototyping at a much lower cost. \nAgenda: \n\nIntroduction to Llama-cpp-python\nInstallation and configuration\nPython API\nOpenAI API\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSeries: \nOver the next several sessions\, we will be diving deeper into separate components needed for RAG – hopefully resulting in a chat-based Q&A service for the NASA Technical Report Server.  We were introduced to this data during our submission for the 2022 NASA SpaceApps Challenge – where we placed 2nd. Our submission was a semantic search based on the abstracts for the NSTR dataset of 10\,000 papers. \nI hope to have the video from our last session posted today. You can check for updates at https://hsv.ai/videos \nLinks: \n\nhttps://github.com/abetlen/llama-cpp-python\nhttps://llama-cpp-python.readthedocs.io/en/latest/\nHuntsville AI 2022 SpaceApps Submission – https://github.com/HSV-AI/spaceapps2022\n\n\nDetails: \n\nDate – 05/01/2024\nTime – 6-7pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/82295484520?pwd=cU1ETGNrS3NtY1hsUmNmRVJDUmNsUT09
URL:https://hsv.ai/event/deep-dive-with-llama-cpp-python/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/04/llama-cpp-python.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240424T180000
DTEND;TZID=America/Chicago:20240424T190000
DTSTAMP:20260405T184846
CREATED:20240129T035257Z
LAST-MODIFIED:20240422T045353Z
UID:1555-1713981600-1713985200@hsv.ai
SUMMARY:2024 SBIR Topics Round 2
DESCRIPTION:SBIR/STTR Topics \nThe second round of SBIR Topics for DoD were released last week. As usual\, we will go through any that are labeled as Machine Learning or Artificial Intelligence – which looks like 13 SBIR and 9 STTR out of 176 total topics. \nYou can find the full list below in the Links\, just search for Learning or Artificial. \n2024 AI Symposium \nThe sessions from the Space and Rocket Center AI Symposium were recorded and just released on YouTube! Based on my notes from the sessions that I was able to attend\, they range from high level overviews to in-depth use of AI. I believe that all of them would be considered practical rather than theoretical. Again – look for the link below. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSeries (picking back up next week): \nOver the next several sessions\, we will be diving deeper into separate components needed for RAG – hopefully resulting in a chat-based Q&A service for the NASA Technical Report Server.  We were introduced to this data during our submission for the 2022 NASA SpaceApps Challenge – where we placed 2nd. Our submission was a semantic search based on the abstracts for the NSTR dataset of 10\,000 papers. \nSessions from this series so far are being posted at https://hsv.ai/videos \nLinks: \n\nDefense SBIR/STTR Portal – https://www.dodsbirsttr.mil/topics-app/\nSpace & Rocket Center AI Symposium Videos – https://www.youtube.com/playlist?list=PLvvHQqQynqmtkQuLsvfFtmy0OohBcA5H4\nHuntsville AI 2022 SpaceApps Submission – https://github.com/HSV-AI/spaceapps2022\n\nDetails: \n\nDate – 04/24/2024\nTime – 6-7pm\nLocation – Zoom\nZoom –https://us02web.zoom.us/j/88996968876?pwd=VzdPTDIzclpxL1FqdHU5cXJPbDVYZz09
URL:https://hsv.ai/event/2024-sbir-topics-round-2/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2023/01/SBIR-Topics.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240417T180000
DTEND;TZID=America/Chicago:20240417T190000
DTSTAMP:20260405T184846
CREATED:20240415T033658Z
LAST-MODIFIED:20240415T033658Z
UID:1627-1713376800-1713380400@hsv.ai
SUMMARY:Choosing an Embedding Model
DESCRIPTION:We have Josh Phillips presenting this week\, so you don’t want to miss this! \nJoin us as we explore the crucial task of selecting optimal embedding models to enhance AI performance across a variety of applications. This meetup will delve into the Multilingual Transferable Embedding Benchmark (MTEB)\, a pivotal resource providing a comprehensive framework to evaluate embedding models over diverse task categories and numerous languages. The selection of the right embedding model is vital\, yet challenging due to the myriad of options and their inherent trade-offs. \nThis presentation will not only introduce you to MTEB’s holistic approach across eight core NLP tasks but will also guide you through the practical steps of identifying\, shortlisting\, and benchmarking models to find the best fit for your specific needs. \nAgenda: \n\nIntroduction to Embedding Models – Gain insights into why choosing the right model is critical for AI tasks.\nOverview of MTEB – Understand the framework of the Multilingual Transferable Embedding Benchmark and its application across 100+ languages.\nDeep Dive into MTEB Tasks – Explore the eight fundamental tasks within MTEB\, including bitext mining\, classification\, clustering\, and more.\nCase Studies – Walk through real-world use cases\, demonstrating how to apply MTEB in selecting models for tasks such as walking path recommendations\, form filling automation\, and building a documentation assistant.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nSeries: \nOver the next several sessions\, we will be diving deeper into separate components needed for RAG – hopefully resulting in a chat-based Q&A service for the NASA Technical Report Server.  We were introduced to this data during our submission for the 2022 NASA SpaceApps Challenge – where we placed 2nd. Our submission was a semantic search based on the abstracts for the NSTR dataset of 10\,000 papers. \nI hope to have the video from our last session posted today. You can check for updates at https://hsv.ai/videos \nLinks: \n\nMultilingual Transferable Embedding Benchmark (MTEB)\nMTEB Github\nMTEB Paper\nHuntsville AI 2022 SpaceApps Submission – https://github.com/HSV-AI/spaceapps2022\n\n\nDetails: \n\nDate – 04/17/2024\nTime – 6-7pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/81626946368?pwd=c2M3QTlXSy9ZS20xdkZrUHBIMHdOdz09
URL:https://hsv.ai/event/choosing-an-embedding-model/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/04/Choosing-an-Embedding-Model.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20240403T180000
DTEND;TZID=America/Chicago:20240403T190000
DTSTAMP:20260405T184846
CREATED:20240402T032141Z
LAST-MODIFIED:20240402T032141Z
UID:1622-1712167200-1712170800@hsv.ai
SUMMARY:Document Chunks with LLM Sherpa
DESCRIPTION:Continuing our discussion about Retrieval Augmented Generation (RAG)\, this week we will incorporate LLM Sherpa to provide chunks of text from PDF documents that have been retrieved from the NASA archive. \nOur initial attempt used PyPDF2 to read text from the PDF documents. It was very slow and provided limited strings of text that did not match the paragraphs in the documents. We’ll take a look back at what was available at the time\, and then look through the LLM Sherpa API and see what it looks like with that piece incorporated. \nAs we get further into this project update\, it has become apparent for the need to split the monolithic application into components that can be hosted and updated separately. We will go through what has been done so far to containerize both the ChromaDB vector database and the LLM Sherpa for chunking.
URL:https://hsv.ai/event/document-chunks-with-llm-sherpa/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://hsv.ai/wp-content/uploads/2024/04/LLM-Sherpa.png
END:VEVENT
END:VCALENDAR