LisA: EXAMPLE

anudha: This is what

Holy: How to query a periodic table? And a nucleotide chart?

Ryan: Price and temperature range for furnaces?

Rick: Where can we get a set of loads to test a battery configuration?

Julie: Configure web servers : caddy / apache / python http server

Tim: Homesteading Drilling https://www.drillawell.com/how-terragrinder-works

Joanna: If I install and configure a web server, all the files of a site (html, css) , all the files related to compute or logic (python, php) inside a docker container, then can I transfer that container to another server and the site would work? Also: https://github.com/dragonflyoss/nydus/blob/master/docs/nydus-design.md

Shree: Can we implement webhooks to get notification on the state of the neural machine?

Damian: SIMD programming model in use today: CUDA, OpenCL, OpenVX, Halide

Hil: I want a new project

Sara: I have a new project.

Subhash: Infra notes on setting up permissions for dirs in a social media platform

Tech Blog: https://engineeringblog.yelp.com/

Text Embedding Model: https://www.sbert.net/docs/sentence_transformer/pretrained_models.html

RNN: https://arxiv.org/pdf/1409.0473 - paper on encoder/decoder and RNN (Recurrent Neural Net)

Oauth: https://datatracker.ietf.org/doc/html/rfc6749

TU Darmstadt: NLP Tasks https://tudatalib.ulb.tu-darmstadt.de/handle/tudatalib/1359

C Lang: https://clang.llvm.org/docs/InternalsManual.html

Kafka: https://kafka-python.readthedocs.io/en/master/

JD from Applied Materials (Most Likely): • 6+ years of IT experience with a min of 3+ years in Data Science (AI/ML) • Strong programming skills in Python • Experience with deep learning frameworks (e.g., TensorFlow, PyTorch) • Hands-on AI/ML modeling experience of complex datasets combined with a strong understanding of the theoretical foundations of AI/ML(Research Oriented). • Expertise in most of the following areas: supervised & unsupervised learning, deep learning, reinforcement learning, federated learning, time series forecasting, Bayesian statistics, and optimization. • Hands-on experience on design, and optimizing LLM, natural language processing (NLP) systems, frameworks, and tools. • Building RAG application independently using available open source LLM models. • Comfortable working in the cloud and high-performance computing environments (e.g., AWS/Azure/GCP, Databricks) Role: AIML Manager / Data Scientist Location: Santa Clara, CA – Onsite opportunity Job Description : •At least 8+ years’ experience, ideally within a Data Science role. •Primary responsibility will be to develop and optimize custom ML/AI algorithms for new and existing applications. •Broad knowledge of computer vision, NLP, time series forecasting, anomaly detection •Knowledge of traditional ML algorithms such as, regression, classification, and clustering algorithms •Knowledge of state-of-the-art deep learning model architectures in the areas of computer vision (NLP would be a plus) •Experience in implementing and optimizing object detection and instance/semantic segmentation models •Experience in setting up end-to-end pipelines for model deployment •Experience in model performance tracking using appropriate KPIs •Strong fundamentals in Python programming •Good knowledge of OpenCV, Scikit-image, TensorFlow, Torch, Pillow, numpy, pandas, scikit-learn etc. •Understanding of SW development cycle, from requirements to testing, integration and delivery Familiarity with model shrinking techniques for deployment on edge devices with limited footprint Nice to have: Experience in process improvement in manufacturing industries using ML/AI Experience in defect identification and root cause analysis in manufacturing domain

JD from Toyota: What you’ll be doing Define research goals & tasks for AI and/or data technology. Work with university researchers on emergent AI and data technologies and algorithms. Research and develop data platforms (open source, commercial) where this platform can be used widely across multiple business services including battery, supply chain, marketing, mobility, chatgpt for support, etc. Innovate and develop new data capabilities. Generate patents and publications. Lead discussions of multiple number of products/applications using emerging data and AI technologies. Research and develop new capabilities within the data domains (data ingest, data integration, data labeling, data cleansing, data processing, data and AI analytics, data management). Develop innovative ideas and solutions for AI and data domains that deliver Better performance and Scalability Lower cost and reduce latency High Availability and Redundancy Observability and Reliability Encryption, Security, etc. Conduct PoC and benchmarks data services/capabilities to compare performance, costs and other criteria. Establish research collaborations with leading experts in universities and industry consortiums Mentoring and training junior researchers in group/department wide Technology transfer ITL research to advanced development or valuable business contribution to the company Obtain patents and publish research in prestigious journals and conferences Regularly champion research and present/influence to VP and GVP level and show AI or data technology expertise. Evangelize the importance of AI, generative AI, AI platform and create vision and roadmap. Requirements: THIS POSITION WILL BE WORKING DIRECTLY AT THE MOUNTAINVIEW SITE DAILY PhD in Computer Science, Math, Statistics or related field Hands on experience in R&D with proven credential in patents and publications - 5 year Strong familiarity with end-to-end data services/domains and technologies. Data Ingest (ETL, ELT, File, Streaming, Queue, etc.) Data integration (Quality, cleansing, labelling, API, etc) Data Processing Data Access Data Analytics (familiar with different tools) Databases (unstructured and structured – SQL, NoSQL, Vector, RDMBS, etc.) Data Storage technology (SSD, HDD, etc) Memory/Cache PoC and Benchmarking experience in AI or data infrastructure Strong research interest in data areas where data is the foundation for AI computer Research experience with data infrastructure, platform development, IoT, etc - 3 years Experience with advanced data platform, distributed cloud systems & architecture including data management, data ingestion, data analytics, data processing, data observability for infrastructure, data normalization, clustering, logs, tools, monitoring, etc. - 3 years Hands-on prototyping and programming knowledge (e.g. Python, SQL/NoSQL, Kafka, Spark, etc). – 3 year Managed research projects, e.g., planning, reporting, and managing time / team / outside collaborators – 3 year Well-developed interpersonal and communication skills, including ability to respond professionally in all types of situation Added bonus if you have: PhD in Engineering (e.g. Computer Science, Data Science, Information/Data Management, Math, Statistics, Information Theory, etc) or related field Corporate lab setting with a proven track record of cross functional collaboration and achievements experience Key components of IoT, edge, distributed cloud architecture with emphasis on connected systems – 3 years Experience of proof-of-concept (PoC) prototyping for connected cars and IoT systems - 3 years Experienced with prototyping and cloud bulk data injestion techniques (e.g. AWS, Spark, Kaftka)- 2 years Programming experience with Python, Rust and embedded systems (e.g. NVIDIA Jetson, Arduino, etc) - 2 years Working experience and knowledge of connect car architecture and vehicle compute requirements - 2 years Broad knowledge of Computer Systems, Software engineering and Embedded Systems Experience working and communicating in multicultural and multinational companies - 2 years

Julia: Using equation (1) in this paper https://arxiv.org/pdf/2007.01852 , each pair in a dataset is either similar or dissimilar. There are only two choices. Levels of similarity are not considered.

SBIR DOE: Make software codes built by government labs easier for public use.

Justin from Palo Alto: Do we really need a web server? Or write a custom one?

Jason: What key is this song in?

Juan: I want to organize similar field / column names into one field name.

SBIR DOE: take one or more ASCR-funded software packages and make them easier to use by a wide variety of industries or in commercial venues by developing commercial offerings based on those ASCR-funded software packages

SBIR DOE: ASCR-funded software packages. SuperLU (https://portal.nersc.gov/project/sparse/superlu/), STRUMPACK (https://portal.nersc.gov/project/sparse/strumpack/), HYPRE (https://www.llnl.gov/casc/hypre/), Trilinos (https://trilinos.github.io/), PETSc (https://www.mcs.anl.gov/petsc/), SUNDIALS (https://computing.llnl.gov/projects/sundials), MFEM (https://mfem.org/). • Programming Models: Kokkos (https://github.com/kokkos/kokkos), RAJA (https://github.com/LLNL/RAJA), Umpire (https://github.com/LLNL/umpire), Legion (https://legion.stanford.edu/) • I/O: ADIOS2 (https://github.com/ornladios/ADIOS2), Parallel NetCDF (https://parallelnetcdf.github.io/), HDF5 (https://www.hdfgroup.org/) • Compilers and Runtimes: LLVM (https://llvm.org/), Argobots (https://www.argobots.org/) • MPI: OpenMPI (https://www.open-mpi.org/), MPICH (https://www.mpich.org/) • Package Management: Spack (https://spack.io/) • Software Stacks and SDKs: E4S (https://e4s-project.github.io/), xSDK (https://xsdk.info/). o Please note that E4S and xSDK include many ASCR-funded software packages that are not separately listed in this document. Return to Table of Contents • Artificial Intelligence: DeepHyper (https://deephyper.github.io/), LBANN (https://github.com/LLNL/lbann) • Software for Quantum Computing and Information: Proposed products stemming from ASCR-funded software and algorithms for quantum information science are in-scope. See https://science.osti.gov/ascr/Research/Quantum-Information-Science-QIS.

Jen: Can we post about materials science problems with home construction? The metal on my railing gets very hot. What material should I look for? I don't want plastic. Wood is okay. Any options in metallic alloys?

Geet: Vents get clogged with chalk very frequently. This is a climbing gym and we have a lot of chalk particles in the air. How do other indoor air pollution facilities deal with this?

Kim from Cupertino : Need to develop a shipping pipeline. Need low cost quotes on packaging materials, adhesives

Kim from Cupertino : Need to develop a shipping pipeline. Need low cost quotes on packaging materials, adhesives

Shawn from Advanced Materials : Measurements from charge - discharge cycles in batteries shows that cells are not discharging completely. Resistance increases before discharge is complete.

Holly: <h1>Submit Questions</h1>

Wayne: <h1> hello <h1/>

Wayne: <h> hello <h/>

Wayne: <h> hello <h/>

Wayne: How to query a periodic table? And a nucleotide chart?

Daniel: How do you keep track of site structure? If I change the file name, I'd run a replace on all the files in the site structure from old_name to new_name. Is there more complexity to changing URLs?

Dayna : Can someone list best practices or methods to reduce inference time for embedding models?

Laura: Which scikit learn classifier should I use?

Shawn: Can we convert Excel to SQL? Will it be faster to search in the data?

Joel: Organizing work to outsource or work with a vendor. Most of our work-flow is linked to the internal web. Data is proprietary. We need to leverage external resources. Organizing and thinking through that.

Sarah: Write APIs for event-driven services.

Nicole: We need AI-ML for chemical vapor deposition processes and etching processes. Increase speed of deposition. Increase speed of etching. Any research papers in this area?

Steven: Which datasets are available for different material characterization instruments? For xrd, sem, tem, stem, optical microscope

Tom R: Log sensor data from different locations in one data file. Send alert for any measurements above threshold limits.

Tristan: Translate documents without errors

Dostoevsky : Recommend new literature

Janet M: We have many tasks to be distributed. Catalog enthalpy values (calorimetric data) and activation energies. Run acid-base titrations. Run mechanical testing strengths. Keep data linked by material / chemical composition and data of data collection. Is there a way we can keep track of storage conditions + deployment conditions? Materials change under pressure/temperature. Keep data live real time. Can people in the field view live data?

Melonie: Summarize all the reports, pdfs, and post it on the news blog everyday. Daily everyone will hear about company historical achievements. Outline relevant problems. Connect with business and revenue.

Kyle: What to do about typos in the data entry?

Ed L: Forecast heating requirements.

Herman: Use NLP (natural language processing) to remove mistakes from the blog posts. Link similar blogs.

Jose: Organize chemical inventory for supplies management and budget. Predict budget for next five years. Share inventory and budget needs with program managers, finance office, facilities space. Also for cold storage needs and notify electrical planning team.

Martin: Create a survey for employees across teams

Martin: Create a survey for employees across teams

Martin: Catalog phase diagrams

Julia: Launch a website to collect ideas internal to my organization

Amanda: Organize data and files accumulated over the years. Create a layout of which excel file and documents are linked to each other. Find data distributed across different spreadsheets. Method to organize vast data: programmatically create an organized chart or visual of all your directories, nested folders, and files. The visual will illuminate work-flows. Use names of columns and rows in spreadsheets and/or excel files to create SQL tables. Spreadsheets lead to easy visualization of data. Easy recording of data in isolated experiments. The next step is to connect different spreadsheets for deeper insights and correlations, planning more complex work. Coming up with a schema that can be searched, i.e. queried with SQL (structured query language) will make data across many years of experiments easier to search. Example from a chemist below: Materials High Density Materials.xlsx Low Density Materials.xlsx Formulations Organic Mechanical_Strengths.xlsx Adhesives.xlsx