Wojciech Maciejewski@Laszer

Revolutionize Document Management for Construction Industry with AI-Powered Solution. Transform unstructured documents into interconnected, actionable knowledge with our cutting-edge Information Retrieval technology. Advanced Document Digitization: • State-of-the-art OCR technology based on OCR4all framework. • Multi-stage process: intelligent preprocessing, advanced segmentation, precision recognition. • Hybrid approach: automated efficiency, manual parameterization, LAREX editor integration. • AI-powered language models for text correction. • Multimodal models (e.g., CLIP) for complex documents. Dependency Analysis: • Named Entity Recognition (NER) for mapping relationships. • Instant insights on responsibilities, contacts, and historical data. Graph Generation: • Interactive visualization of company knowledge. • Explore document dependencies, semantic relationships, and personnel responsibilities. • Hybrid search technology combining dense and sparse methodologies. • Intuitive exploration tool. • DAO for node traversal, ready for LLM-chat integration. Business Benefits: • Structured, searchable knowledge base. • Uncover hidden insights. • Scalable information management. • Secure, persistent storage with automated backups. • Improved employee focus on high-value tasks. • Find discrepancies across all your documents and address them before they cause project-wise problems. Advanced Features: • LLM-powered chat application integration. • Choose from Azure, Gemini, or Groq. • On-premises LLM server option for enhanced security. Security and Compliance: • Offline processing is possible on your infrastructure. • Meets stringent security and compliance requirements. Transform your document management approach and unlock the full potential of your information assets.

Our project introduces a groundbreaking AI-powered Gemini-based chatbot designed to revolutionize the e-commerce experience, dramatically increasing conversion rates and reducing cart abandonment. By leveraging state-of-the-art technologies such as Milvus vectorstore for scalable data retrieval, Gemini LLM as an LLM agent, and the OpenAI Ada embeddings for advanced product recommendations, our chatbot mirrors the capabilities of an in-store assistant online. It offers personalized product suggestions, detailed comparisons, and technical specifications, along with handling customer service inquiries efficiently. We focus on building a customer-centric, specialistic solution to solve real-life problems. For that we leverage techniques such as continual learning and adaptability to each customer's needs through few-shot learning and dynamic response templates. This solution not only enhances the customer shopping experience but also provides invaluable insights into customer behavior, leading to increased average order values, sales growth, and improved SEO. With a focus on AI safety and the possibility of on-premise model hosting, our approach is both innovative and secure. Our team brings expertise in LLMs, AI, ML, and web development, ensuring a comprehensive, customer-centric approach to e-commerce challenges.

Using StableCode-Completion model, we have built a LaTeX Copilot that can help scientists and engineers around the world write their LaTeX documents (papers, articles, documentation) more easily, faster, and with more enjoyment. We plan on replicating the success GitHub Copilot had with programmers community. After extensive research, we know what we want to make and how. The technology is tested, the plan is laid down, resources are acquired and waiting. Our solution is not only a viable business product, we also care about broader impact. We believe that our product can expedite the process of scientific research on various important topics changing the world for the better. If you want to help us in changing the world, then please contact us!

Times when you had to do multiple takes while recording your presentation are over! With Vocaly, you can enhance audio and speech quality, use your best voice in all your presentations, get rid of all recorded stuttering and similar defects, and even edit already recorded presentation simply by editing the text! Because of those features, Vocaly provides a solution to those with speech impediments or Tourette syndrome. Our solution will also come in handy for any person who isn't a professional presenter, especially when they present in their non-native language and struggle with pronunciation. Vocaly let's you do all that and even more. To fuel even higher inclusion we also enable our users to add automatic subtitle to their videos and even translate the whole speech in a recorded presentation. You can present in your own language and then translate that to another language like English. Then you can correct all mistakes made by the AI translator to really polish your presentation. Vocaly uses elevenlabs for voice generating and voice cloning, pvleopard library for speech-to-text and openai's GPT-3.5 for imputing punctuation. All of that is presented to a user by using a clean and elegant frontend in React. All in all, we are really proud of how this application turned out. It works well enough, even as a prototype, that we have actually used it for editing our presentation on lablab.

Job hunting is always hard and tedious work. What if someone could do it for you? Do you want to fill in your information and preferences only once and then see the mails from recruiters rolling in? Even better, your information could be pre-filled based on your CV and updated by following updates on your LinkedIn profile. If you said yes, then we have something just for you. Something that will let you forget about searching for a job while still constantly getting new calls from recruiters. That's our aim, to automate job hunting. Currently the MVP works with LangChain, ChatGPT (GPT-3) and Chroma vectorstore for handling complicated NLP tasks. We are also using Selenium for automatically filling forms on a website hosting job offers. New features such as creating and updating your CV, making you creative responses for cover letters and other open-ended tasks in the forms are already being planned. Obviously, right now we have only a prototype but if you are interested about this unique opportunities then feel free to contact us on social medias, we will keep you updated :)

Writing scientific articles goes brrrrrr... with our newest copilot application for LaTeX called LatexAI. LatexAI is powered by Google's PaLM and is focusing on helping the writing process of technical and scientific articles. During ideation we have identified 3 core functionalities that could help our users boost both effectiveness and efficiency of their work. Those functionalities are: 1. Autocompletion working both for natural language (text) and LaTeX syntax. 2. Suggestions for citations and assistance in creating a bibliography. 3. Conducting a comprehensive review of the paper. We believe that by providing and improving those 3 functionalities we can provide a lot of value to our users. There is also a lot of things planned for further development but what we want to focus on is rolling our product as beta and acquiring users' feedback as soon as possible. Then our further development will be driven by addressing our users' feedback.

Personal Elf is an AI-driven application for recommending gifts. Have you ever struggled with choosing the most suitable gift for one of your close ones? Don't worry, from now on your Personal Elf has your back! This application is driven by OpenAI's models. We used GPT-like transformer-based DaVinci model for generating gifts proposals and then visualized proposed gifts with diffusion model called DALL-E 2. We created a demo using streamlit that was also deployed using their service. We highly value inclusivity, therefore we designed our solution to be convenient also for other occasions, such as birthdays, anniversaries or hanukkah. Such flexibility allows us to make our product more accessible to users and it can be advertised accordingly during the year. While building our userbase and being able to drift afloat thanks to adverts, we want to get into partnerships with various online shops. We believe we can provide them with a great advertisement by integrating their shops into our application and recommending the users to buy on our partners' sites.

storytime.ai is a innovative, creative application of stable diffusion model (dreamstudio.ai) paired with transformer model(cohere.ai). We leverage powerful features of those models in attempt to solve vital problem: storytelling. It's important to recognize crucial drive of humans to create stories. Now, with help of artificial intelligence our stories could get even more creative and picturesque. Think of all the possible aplications of on-demand stories generator. Batteries (and stylish ilustrations) included! Our solution allows parents to tell brand-new fairy tales to their children. Writters, marketing specialists, basically any proffesion or bussines that deals with creating stories would also benefit from storytime.ai

🗓️ This will be a 48-Hour Virtual Hackathon from 2 - 4 December 💻 Technology: You will build applications with generative AI supplied by Cohere ✔️ Level: All levels are welcome 😊 For whom?: Builders, creators & innovators! 💲 The Event is totally free! 🏆 $2000 cash prize pool!

🗓️ This will be a 7-days of hacking and fun from 9-16 December 💻 Build with the latest AI tools from OpenAI to create innovative new apps 💡 Work with top AI professionals and learn from them ⚒️ Create your AI app by combining GPT-3, Codex, Dalle-2, and Whisper 🐱‍💻 Register now and let's get started!

🗓️ This will be a 7-day virtual hackathon from 16-23 December 💻 Build AI application with the latest large language model-powered technology by Cohere 💡 Get the chance to work with the best AI professionals in the industry and learn from them ✔️Entry level = 0. You’ve just started with AI? Are you an experienced Data Scientist? Or maybe you are a Designer or Business Developer? Join us! We need your domain knowledge! 🐱‍💻 Register now and let's get started! It’s free!