DISQO is the CX intelligence platform for understanding every customer experience

DISQO is the CX platform for understanding every customer experience. Businesses trust DISQO to power better decisions for every customer, touchpoint, and outcome. DISQO’s insights, agile testing and advertising measurement products are powered by millions of consumers on the industry’s largest opt-in consumer data platform. DISQO is recognized in Deloitte’s Fast 500 and Ad Age’s Best Places to Work, and is a Top 100 Next Gen Workplace.

DISQO

Real-world cases and expert opinions about the present and future of audio deepfakes and AI-powered fraud in particular and how to use AI to build a strong defense against malicious AI. John Dow, the CEO of an unnamed UK-based energy firm, once from his boss that he wishes he'd never answered. Being confident he was talking to the CEO of the firm's German parent company, he followed the instruction to immediately transfer €220,000 (app. $243,000) to a Hungarian supplier's bank account. got a call Having completed a transaction, John got another call from the boss confirming the reimbursement. However, he noticed that the purported reimbursement hadn't gone through, and the call had been made from an Austrian phone number. At this point, John got the third call seeking a follow-up payment. That's where he went sceptical and sounded the alarm. The investigation found that the voice in the call belonged to a hacker using AI voice technology to spoof John's boss. In fact, the fake voice was so close to the real one that John could recognize the subtle German accent and the man's voice "melody". Audio deepfakes are gaining traction and that's creepy According to , audio deepfakes are extremely hard to catch red-handed. Victims rarely have any evidence as they don't typically record inbound calls, and the technology detecting synthetic speech in real-time is still in its infancy. As a result, the actual number of cyber fraud cases based on audio deepfakes remains very low. According to antivirus provider Symantec, in 2020, there were only three registered cases where audio deepfakes were used against their clients. Forbes "Applying machine learning technology to spoof voices makes cybercrime easier," says Irakli Beridze, head of the Center on AI and Robotics at the UN Interregional Crime and Justice Research Institute. I've reached out to , co-founder and CPO at , an identity verification and compliance company, for a comment on the future of deepfakes. Jacob Sever Sumsub " It still takes plenty of effort to make a truly realistic deepfake. But I believe that, in the future, creating them will take just a few clicks. Indeed, the deepfakes of the future will be so high-quality that you won't be able to distinguish them from a real individual. Once deepfakes can be made easily, we'll be faced with a game of cat-and-mouse. That's because the only way to fight malicious deepfakes would be through the same neural networks that create them. So, as deepfake and neural network technology become more sophisticated, we'll have to upgrade our defences constantly. " AI will eventually learn to recognize any deepfake through analyzing specific, invisible features. Nevertheless, we'll still need to fortify identity verification going forward. This includes implementing technology based on state-of-the-art cryptography, including chips that could be embedded in humans for identification. Another plausible scenario of malicious AI usage that can be detrimental for business is . robocalling If you get a call, answer it and hear a recorded message instead of a live person, it's a robocall. If you get many robocalls trying to sell you something, such calls are most likely illegal and are scams. According to the , robocalling is one of the biggest consumer complaints they receive daily. Federal Communications Commission How does robocalling work? Scammers use AI and ML to fully automate their robocalling systems and discover types of victims that are likely to respond to their fraud or better understand which logic and arguments would lead to more successful shenanigans. With an abundance of data points, cybercriminals refine their scamming schemes and wage more advanced and sophisticated attacks. I've talked to Nelson Sherwin, General Manager of that provide HR and regulation compliance services to businesses. The company fell victim to a flurry of autodialer calls early in 2020. PEO Companies " Our old telephony system was one we somewhat inherited with our space, and it was working fine for us until this. There was no real way for our IT department to manage the influx of calls adequately. It very quickly began to impact my employees' capability to do business with our clients. Employees essentially had to disconnect their phones for several hours over a few days while working out a solution. We hastily but successfully switched to a software VoIP solution, which allowed our IT team to use a much better call management environment while providing the granularity needed to block the incoming calls much quicker. ." Eventually, after what I assume were hundreds of repeated attempts that failed, they gave up on trying to bombard our phones again How organizations use AI and ML to fight and prevent AI-driven cyber fraud and deepfakes Ragavan Thurairatnam, co-founder and Chief of Machine Learning for , a startup currently working on , believes using a traditional software-based approach would make it extremely difficult to figure out the rules to write in order to catch deepfakes. Dessa developing AI for detecting audio deepfakes "On top of this, deepfake technology will constantly change, and traditional software would have to be rewritten by hand every time," Thurairatnam. "AI, on the other hand, can learn to detect deepfakes on its own as long as you have enough data. In addition, it can adapt to new deepfake techniques as they surface even when detection is difficult to human eyes." explains To build public awareness about the risks of AI-powered synthetic speech, a few months ago, Dessa shared their own example of the technology with the broader audience. Using a proprietary speech synthesis model, they built RealTalk, a recreation of the popular podcaster Joe Rogan's voice. How deepfake detector works To distinguish between real and fake audio, the visual representations of spectrograms, which are also used to train speech synthesis models. While sounding almost identical to the unsuspecting ear, spectrograms of fake versus real audio differ. detector technology uses Engineers at Dessa trained their detector on Google's 2019 containing over 25,000 audio clips featuring fake and real voices of a great variety of male and female speakers. AVSSpoof dataset The company built a deep neural network that uses Temporal convolution. Raw audio is pre-processed and then converted into a mel-frequency spectrogram, which serves as the input for the model. The model performs convolutions and uses masked pooling to prevent overfitting. The output is finally passed into a dense layer and a sigmoid activation function, which ultimately outputs a predicted probability between 0 (fake) and 1 (real). Image source: Dessa As a result, the detector model can currently predict over 90% of the audio deepfakes, which puts it front and center in the effective combat of AI fraud. Building predictive flags for AI fraud potential When an enterprise creates insights for decision-making on fraudulent data, the consequences can be dire — product design misses the mark, while faulty competitive intelligence or bad brand experiences damage reputation and user loyalty. This inspires companies having the firsthand experience with AI-driven fraud to develop defensive technologies. One example is , an audience insights platform that has modernized the panel-based approach to consumer surveys. In particular, the company employed Cloud-based technology allowing users to easily participate in research anywhere on any device. DISQO asks people if they would be willing to share their opinions and online behaviors, and rewards them for this. DISQO "This is the antithesis of prevalent ad tech solutions that use technology to deduce consumer identity and follow them online without their full understanding or consent," says , DISQO's Chief Customer Officer. Bonnie Breslauer DISQO's approach has skyrocketed in popularity, with more than 10M users to date and more than $25M in cash rewards paid out. But wherever money is to be made, thieves are sure to follow. Driven by the allure of the compensation for survey participation and likely spurred by the pandemic, fraudsters generated a huge spike in fraudulent activity in 2020. This inspired DISQO to lead in fighting fraud by developing defensive technologies that keep cheaters out of their panel. DISQO's data scientists built predictive flags for fraud potential by studying good and bad interactions with their platform. As they needed more computational power, they moved to Spark Engine, which gave the team the needed "horsepower" for next-level machine learning. DISQO created a scoring system to rank panelists' behaviors that might indicate fraudulent activity. Using Bayesian optimization, the simple initial grading system was evolved into a model for maximum fraud detection, which continuously improves through ML. In less than six months, DISQO's data science team achieved its best results by solving the optimal fraud coverage segment and training the company's machine learning models on a hand-picked number of classifiers. The team constructed a fraud vector of four dimensions that consisted of the probabilities retrieved from the hand-picked classifiers. One of the fraud vector's dimensions contained the CNN probability; the other two were XGBoost and Logistic based probabilities. The system learned the separation in the data between "good" and "bad" panelists. DISQO now compares fraud vector magnitudes on every user to assess overall fraud risk. This allows DISQO to detect and remove "bad actors" while preserving a high-quality experience for the rest of the community. Final Thoughts , VP of AI/ML at and the former Principal at Microsoft AI and Research Group,  believes there's nothing inherently malicious about AI. Shane Lewin GSK " Facebook currently runs an AI-powered feed that is constantly learning how to keep you engaged on the platform. It exploits parts of your brain that reinforce community and social structure with the intent for more screen time. Those AI models are quite effective but not without cost — a convincing and overwhelming body of evidence suggests it's quite harmful to your productivity, relationships, happiness, and even health. I've not personally been subjected to deepfakes, but I've seen the impact firsthand — a targeted and deliberate deepfake campaign leads to real reputation and health impacts to an innocent victim. Racial and gender bias is rampant in the search algorithms on career pages, notably Linkedin. This is actually not malicious in intent, but the AI has learned the behavior of the bias in users, based on their clicks; or the biases of the people who trained it. There's nothing inherently malicious about AI — it's just software that gets iteratively better at some defined function. The analogy is a human businessperson trying to make more money. They might try a wide variety of strategies and increase usage and investment in what works, but our value system constrains most humans. There're lines we won't cross – things that are too odious to profit from. These rules — we call them values — limit our profit to a degree, but society is better for having them. Alas, not all humans are self constrained. There's a class of us that we have names for, ranging from selfish to narcissistic to psychopaths, but that all describes individuals that are too lax with our values. Society has learned to reject these people, to a degree at least. " AI is the same. If it's not constrained with some value system, it will adopt whatever strategy is optimal without any sense of values or ethics. The Facebook algorithm says maximize screen time when it should say maximize screen time within constraints above, which are dangerous to mental health (or better yet, just turn the personalized feed off). Linkedin should maximize relevance without any signals that might convey bias (such as user click data or racial identifiers). It's quite possible to build ethical AI if you're willing to put values above profit. I can't agree more! There are many malicious companies, led by malicious executives, deliberately maximizing profit in an evil way. And to date, there's no better way to fight AI-based fraud than using AI. Fight fire with fire! Sharing and liking this article is appreciated! If you have an interesting tech story that you believe deserves a good shoutout – just drop me a line on LinkedIn and I'll be happy to cover it in my next article on Hacker Noon.

Fight Fire with Fire: Using Good AI to Combat Bad AI

The world of work is changing––rapidly. Thanks to the advent and rapid adoption of generative AI, tech stocks had a bumper 2023, with the share price of the Big Five ( , , , , and ) leaping by 60% from the start of that year. 2023 was also a , which saw its shares triple in price. Alphabet Amazon Apple Meta Microsoft smash success for chipmaker, Nvidia And while adoption isn’t even across the board yet, corporate America is sitting up and taking notice. At JPMorgan Chase, for example, the company said it had 600 machine-learning engineers, with AI working on more than 300 different internal applications. A recent found that 71% of C-suite leaders said that advancing their use of AI is either a high or medium priority for their organization, with 62% of executives noting that they are worried they’re not moving fast enough with AI. study from UKG That has led to some companies doing “AI washing”, aka exaggerating the amount of AI technology they use in their products or services. While some companies might not be entirely truthful about their motivations and use of artificial intelligence or machine learning technologies, what is increasingly true is that in an AI future, tech workers will need to upskill. The World Economic Forum (WEF) says that “the 2020s are becoming the decade of development for employees and employers in terms of skills and career progression”. Its most recent identified that 44% of workers’ skills will be disrupted in the next five years, and six in 10 staff members will require training before 2027. Future of Jobs Report Upskilling for free Key skills to acquire will include new technology capabilities, but the WEF has also identified key human skills such as analytical and creative thinking, as well as resilience, flexibility and agility, motivation, and self-awareness. AKA, the things that artificial intelligence (so far) can’t do. If you’re thinking about some upskilling, then one great place to begin is (OCW) platform. MIT’s OpenCourseWare A free and open space to learn, OCW contains material from thousands of MIT courses across every MIT department and degree program, ranging from the introductory to the most advanced graduate level. Courses include a syllabus, instructional material (such as lecture notes or a reading list), and learning activities like assignments or exams. You can use all the courses at your own pace, and there is no enrollment fee or date to complete by––useful for those working in busy, pressurized roles. A huge range of computer science, engineering, and technology courses are available, across topics such as machine learning, generative AI, Python, deep learning, cognitive robotics, data science, and much more, all for free. If upskilling to get a new job is on your 2024 agenda, then the courses above are a great start. And your job hunt is made far easier too thanks to the . It contains thousands of interesting tech roles, like the three below. HackerNoon Job Board DataAnnotation is seeking a to train its AI chatbots to code. This flexible, part-time or full-time role allows you to choose the projects you want to work on, involves coming up with diverse problems and solutions for a coding chatbot, writing high-quality answers and code snippets, and evaluating the code quality produced by the AI model. You will need to be proficient in at least one programming language and able to solve coding problems (LeetCode, HackerRank). Software Developer Adobe is looking for a in San Jose. The salary range for this position is $135,200 - $250,900, and you will have either research or industry experience in building scalable data and ML platforms. In this role, you will have the opportunity to build a new GenAI foundation to change the creative landscape and will collaborate with data platform engineers and architects to seamlessly integrate low latency data pipelines into the ML platform for model training, as well as writing high quality, product level code that is easy to maintain and test following standard methodologies. Other tasks include designing and implementing reusable and scalable data-loading framework to support video and audio foundational model training in large-scale and distributed environments. Machine Learning Engineer Python programmers looking for a new role can consider this role at SAIC in Chantilly. You’ll be experienced, results-oriented, and mission-driven, with a specialized focus on data engineering in order to perform data model design, data formatting, and ETL development optimized for efficient storage, access, and computation in support of national security objectives. A solid understanding and experience with developing, automating, and enhancing all parts of the data pipeline to include ingestion, processing, storage, and exposing data for consumption is required, along with a focus on improving inefficient tooling and adopting new transformative technologies while maintaining operational continuity. Python Developer Want a great new tech role? Discover it now on the HackerNoon Job Board right By Kirstie McDermott

MIT Has Released a Bunch of Free Courses: Discover 5 To Boost Your Tech Career

এমআইটি একগুচ্ছ বিনামূল্যের কোর্স প্রকাশ করেছে: আপনার কারিগরি ক্যারিয়ারকে উন্নত করতে 5 আবিষ্কার করুন

Das MIT hat eine Reihe kostenloser Kurse veröffentlicht: Entdecken Sie 5, um Ihre Tech-Karriere voranzutreiben

El MIT ha lanzado una serie de cursos gratuitos: descubra 5 para impulsar su carrera tecnológica

Le MIT a publié un certain nombre de cours gratuits : découvrez-en 5 pour booster votre carrière technologique

एमआईटी ने कई मुफ्त पाठ्यक्रम जारी किए हैं: अपने तकनीकी करियर को बढ़ावा देने के लिए 5 पाठ्यक्रमों के बारे में जानें

MIT が無料コースを多数公開: テクノロジーのキャリアを伸ばす 5 つのコースをご紹介します

MIT가 다양한 무료 강좌를 출시했습니다: 기술 경력을 향상하는 5가지를 찾아보세요

O MIT lançou vários cursos gratuitos: descubra 5 para impulsionar sua carreira tecnológica

MIT выпустил несколько бесплатных курсов: откройте для себя 5, которые помогут улучшить вашу техническую карьеру

MIT Bir Çok Ücretsiz Kurs Yayımladı: Teknoloji Kariyerinizi Yükseltecek 5 Kursu Keşfedin

MIT đã phát hành một loạt khóa học miễn phí: Khám phá 5 khóa học để thúc đẩy sự nghiệp công nghệ của bạn

麻省理工学院发布了一系列免费课程：探索 5 门课程，助你提升科技职业生涯

Finding the right AI model to build a workflow around is hard. With so many models available across different platforms, it’s impossible to know where to start or how to find the one that best fits your specific needs. That’s the problem I set out to solve with AIModels.fyi, the only search engine for discovering and comparing AI models from across the web. The site has a simple premise: describe your problem in natural language, and get back a list of AI models that can potentially solve it. For example, you might search for “I need a model that can help upscale my images without losing clarity” or “make my cat look like it’s singing”. The goal is to , not just ML engineers, to find useful AI tools for their projects. make it easy for anyone Under the hood, the search works by taking the user’s query and comparing it to a database of over 240,000 AI models from 7,500+ creators across major platforms like Hugging Face, Replicate, Cerebrium, and DeepInfra. When a user types in a search, the query gets embedded into a vector using OpenAI’s text-embedding-ada-002 model. This query vector is then compared to pre-computed vector representations of the model names, descriptions, and use cases in the database. The most similar models by cosine distance are returned as the search results. Here’s a simplified version of the core search logic I’ve been using for about six months: const fetchData = async (query) => {
  // Embed the search query
  const embeddingResponse = await openAi.createEmbedding({
    model: "text-embedding-ada-002",
    input: query,
  });
  
  const embedding = embeddingResponse.data.data[0].embedding;

// Find the most similar models in the vector database 
  const { data: modelsData, error } = await supabase.rpc("search_models", {
    query_embedding: embedding,
    similarity_threshold: 0.75,
    match_count: 10,
  });
  return modelsData;
}; This vector search approach works reasonably well for surfacing relevant models when there’s a direct overlap between the query terms and the model descriptions. However, it struggles with more abstract, complex, or niche queries where the user’s language doesn’t exactly match the technical terminology in the database. For example, I recently wanted to find a model that could animate a portrait image using an audio voice clip to create a talking head effect. I knew there were some good models out there for this, but my searches for things like “talking portrait from image and audio” weren’t turning up the results I expected. The problem was that my query terms didn’t have enough literal matches with the specific words used in the model descriptions, even though the semantic intent was similar. This got me thinking about ways to improve the search experience to better handle those kinds of abstract or complex queries. How could I make the system smarter at understanding the user’s true intent and mapping it to relevant models in the database? That’s when I came across the idea of Hypothetical Document Embeddings (HyDE) from after reading about it in . this research paper this Medium post Meet Mr. HyDE — The query expander! The key insight behind HyDE is to use a large language model (LLM) to expand the user’s short query into a more detailed hypothetical “ideal” document before embedding it for search. So, instead of directly matching the query embedding to the database embeddings, it first translates the query into a richer representation in the same semantic space as the real documents. Here’s how the process works: The user enters a natural language query expressing their needs, like “I want to convert an image into a 3D model.” This generated pseudo-document is then embedded into a dense vector using a model like OpenAI’s text-embedding-ada-002. The resulting embedding is used to query the vector database and find the most similar real model documents. The query is fed into an instruction-tuned LLM like GPT-3 to generate a hypothetical document. The prompt asks the model to imagine an ideal document for answering the query. For example: Query: I want to convert an image into a 3D model Hypothetical Document: There are several AI models available that can convert a 2D image into a 3D model. The process is known as single-view reconstruction or 3D estimation. It typically involves using deep learning to predict the depth map and 3D shape from the 2D input image. Some popular models for this task include Pix2Vox, Pixel2Mesh, AtlasNet, and OccNet. These models are trained on large datasets of 3D objects and can handle complex shapes. To use them, you provide an input image and the model will output a 3D mesh or voxel representation that you can render or edit further in 3D modeling tools… The LLM step helps bridge the gap between the user’s simple query and the more technical language used in the model descriptions. It expands on the initial query to infer additional relevant terms, concepts, and context that can help retrieve better matches. At the same time, using real document embeddings as the final search index keeps the results grounded and prevents hallucination. I implemented a basic version of HyDE on top of the existing AIModels search, and the initial results are promising. In some sample queries I tested, HyDE surfaced relevant models that the original keyword-matching approach missed. For the talking head portrait query, it was able to generate a hypothetical document with key terms like “lip sync”, “facial animation”, and “audio-driven animation”, which helped retrieve some strong model matches that the plain search had overlooked. two to three times more How to implement HyDE in your own search system Let me show you how HyDE can be added in on top of an existing semantic search system to dramatically improve its quality. Let’s start with a simplified code sketch based on my current search flow. Here’s the full file — you can glance through it in totality before we look at the individual parts: const fetchData = async (query) => {
  // Generate a hypothetical document embedding for the query 
  const hypotheticalDocument = await generateHypotheticalDocument(query);
  
  const embeddingResponse = await openAi.createEmbedding({
    model: "text-embedding-ada-002",
    input: hypotheticalDocument,
  });
  
  const embedding = embeddingResponse.data.data[0].embedding;

// Search the real model embeddings using the hypothetical document embedding
  const { data: modelsData, error } = await supabase.rpc("search_models", {
    query_embedding: embedding,
    similarity_threshold: 0.75,
    match_count: 10,
  });
  return modelsData;
};
const generateHypotheticalDocument = async (query) => {
  // Use GPT-3 to expand the query into a hypothetical ideal document
  const prompt = `The user entered the following search query: 
  "${query}"
  
  Please generate a detailed hypothetical document that would be highly relevant to answering this query. The document should use more technical language and expand on the key aspects of the query.
  Hypothetical Document:`;
  const response = await openAi.createCompletion({
    model: "text-davinci-002",
    prompt: prompt,
    max_tokens: 200,
    n: 1,
    stop: null,
    temperature: 0.5,
  });
  const hypotheticalDocument = response.data.choices[0].text.trim();
  return hypotheticalDocument;
} Let’s see how it works. We’ll start with the function, which is the main entry point for the search process. First up: fetchData const hypotheticalDocument = await generateHypotheticalDocument(query); This line calls the function (which we'll explain in more detail later) to generate a hypothetical ideal document based on the user's search query. generateHypotheticalDocument The hypothetical document is a key part of the HyDE approach, as it aims to capture the user’s search intent in a more detailed and technical way than the original query. const embeddingResponse = await openAi.createEmbedding({
  model: "text-embedding-ada-002",
  input: hypotheticalDocument,
});
const embedding = embeddingResponse.data.data[0].embedding; These lines use OpenAI’s model to generate an embedding vector for the hypothetical document. I think it's the best embedding option out there so far. text-embedding-ada-002 Embeddings are a way to represent text as a dense vector of numbers, where similar texts have similar vectors. This allows us to perform semantic similarity searches. By embedding the hypothetical document instead of the original query, we aim to get a vector representation that better captures the expanded search intent. Next up: const { data: modelsData, error } = await supabase.rpc("search_models", {
  query_embedding: embedding,
  similarity_threshold: 0.75,
  match_count: 10,
}); This part performs the actual similarity search in the model database. It is called a remote procedure ( ) on the Supabase database, passing in the embedding of the hypothetical document as the . search_models query_embedding The and parameters control how closely the results need to match and how many to return. similarity_threshold match_count The search looks for model embeddings that are most similar to the query embedding based on cosine similarity. Now let’s look at the function: generateHypotheticalDocument const prompt = `The user entered the following search query: 
"${query}"

Please generate a detailed hypothetical document that would be highly relevant to answering this query. The document should use more technical language and expand on the key aspects of the query.
Hypothetical Document:`; This part constructs the prompt that will be sent to the language model to generate the hypothetical document. The prompt includes the user’s original query and instructions for the model to generate a detailed, technical document that expands on the query’s key aspects. Crafting an effective prompt is crucial for guiding the language model to generate useful hypothetical documents. Now for the next part: const response = await openAi.createCompletion({
  model: "text-davinci-002",
  prompt: prompt,
  max_tokens: 200,
  n: 1,
  stop: null,
  temperature: 0.5,
}); These lines use OpenAI’s model to generate the hypothetical document based on the prompt. text-davinci-002 The parameter limits the length of the generated document while controls the randomness (higher values make the output more diverse but potentially less focused). Adjusting these parameters can help tune the quality and variety of the generated documents. max_tokens temperature Next, we have: const hypotheticalDocument = response.data.choices[0].text.trim(); This line extracts the generated hypothetical document text from the GPT-3 API response. The generated text is then returned to be used in the embedding step. Then we can just retreive the data as before, this time using the hypothetical document as the query, but getting a lot more relevant results due to the expanded search term and all it’s extra, juicy context! Recap — what did we just build? I hope the code and my explanation demonstrate how the HyDE approach can be implemented on top of an existing semantic search system. It’s not that hard! The key idea is this: By inserting the hypothetical document generation and embedding steps into the search pipeline, we aim to improve the search results by better capturing the user’s intent. To recap, the key parts were: Using a language model to expand the user’s query into a more detailed, technical hypothetical document Embedding the hypothetical document to get a vector representation that aligns with the model database Performing a similarity search between the hypothetical document embedding and the model embeddings to find the most relevant results Of course, this is just a simplified implementation (simplementation?), and there are many details and optimizations to consider in a real implementation. But it illustrates the core flow and components of the HyDE approach. Taking it further Since this is just an initial prototype there’s still a lot of room for experimentation and refinement. Some of the key open questions I’m thinking about: What’s the optimal prompt structure and output length for generating useful hypothetical documents? I’d like to try different prompt templates and output sizes to see how they impact result quality. How can the hypothetical documents be made more interpretable and transparent to users? It could be confusing if users don’t understand why certain terms are being added. I’m considering showing parts of the generated text as a “query interpretation” to provide more insight into the process. To what extent does the result quality depend on the specific LLM and embedding models used? I’m curious to benchmark different model combinations (e.g. GPT-4 vs GPT-3, vs OpenAI embeddings, etc.) to find the best recipe. Contriever How can this approach be efficiently scaled to work over millions of database entries? Inference on large LLMs like GPT-3 is expensive and slow. I’ll need to look into caching, optimizing, or distilling the models to make this viable at scale. I’m excited about the potential for language models to help make search interfaces like AIModels more intuitive and powerful for a wider audience. Bridging the semantic gap between how users naturally express their needs and how AI models are technically described feels like an important piece of the puzzle. Techniques like HyDE point a way toward search experiences that deeply understand the intent and meet users where they are. My long-term goal is to keep refining this interface until anyone can easily find relevant AI models for their needs, no matter their level of technical expertise. I’ll be sharing more details as I continue experimenting with HyDE and other approaches. If you’re curious to learn more about the technical side, I recommend checking out the original paper. And if you have ideas or feedback on the AIModels search experience, I’d love to hear from you! Let me know what you think in the comments. Also published . here

Improve Semantic Search and Enhance AI Model Discovery with Hypothetical Document Embeddings (HyDE)

I worked in research all my life, so I know a stereotype that researchers write ugly code (e.g. see , , or ). But I thought: we can fix it, right? So multiple times I tried to design nice research frameworks. I tried to bring in interfaces and create nice abstractions using software engineering books and blogs I liked reading. here here here But over and over again all those efforts went in vain. The majority of research software I worked on never went to production (although some did). It would have been great for my mental health if someone told me a simple truth: . Researchers should not spend much time engineering it in the first place. dying research code is actually what is supposed to happen Professional software engineers always look down on researchers who are not using the best software practices. There are several posts trying to raise the bar of research code (e.g. this and a ). But this post goes other way around: it argues how to not overdo the best software practices and instead invest only in fast exploration. It is targeted for research-oriented companies where your goal is to try out many ideas fast. great post research code handbook 1. Take on some strategic tech debt A successful research project at a company has two phases: exploration and exploitation. In “exploration” you want to try out as many diverse solutions as you can. During “exploitation” you have to robustify the best solution and turn it into a useful product. Optimal software practices are quite different between the two. That's why companies often have separate research and product divisions. All the books you might typically read on software design are mainly about the second "exploitation" phase. In this phase you are building foundations for a scalable product. This is where all the design patterns come in: nice APIs, logging, error handling and so on. But in the first "exploration" phase you are building foundations that will live forever. In fact, if the majority of your efforts survive, then you (by definition) did not explore enough. not Many practices in this post are examples of what would normally become “tech debt”. It’s what you get by not writing clean reusable well-abstracted code. Is debt always bad? We prefer never getting a loan or a mortgage, but borrowing money is often a good strategy in life. It's ok to get into debt to move fast and profit later. Similarly, by not taking technical debt you might be slowing down your research. The good news is that the majority of the time you don’t have to pay it back. Most of your research code is likely to die anyway. So on average, you will not be suffering from the whole tech debt you've taken. The case against code reuse Many software architectures and refactoring techniques are specifically oriented to improve code reusability. There are generic to code reuse. But in production they are outweighed by the well-known benefits (for example, see ). In research projects, the majority of the code is destined to sink to oblivion. Striving for code reuse could actually slow you down. downsides this typical post Limiting code reuse is the type of technical debt that is ok to take in research. There are several patterns of code reuse I want to discuss: adding an unneeded dependency, copypasting code, maintaining a lot of shared research code, premature design investments. Think twice before importing something new If you know a well-maintained versioned library that is going to speed you up - go for it! But before taking in a new dependency, try to make a judgment call whether it’s worth it. Every additional one brings you closer to dependency hell. It makes you invest time into learning and troubleshooting it. See more pitfalls of dependencies in this . concise post It’s probably fine to depend on something if: you used it already, there is not much to learn, it has a large community, good docs and tests it is versioned, easy to install and finally, there is no way you can implement it yourself. But be wary about a dependency if: you can’t figure out how to use it quickly, it is very new (or very old) or no one seems to know about it; there are no docs or tests it is from your monorepo and is constantly being changed by other teams it pulls in many other dependencies and tools; or it's just hard to install and finally, you feel that you (or some LLM) can write this code in a few hours. Instead of an explicit dependency, you can follow a nice Go proverb: “ ", which is our next topic. a little copying is better than a little dependency Copypaste gives you freedom of experimentation Some say that “ ”. But to my surprise I found myself arguing in favor of it quite often. Copypaste could be the optimal choice during the exploration phase. copy-paste should be illegal If you depend on a heavily-used function from another part of the codebase you can forget about easily changing it. You’re likely to break something for someone and have to spend precious time in code reviews and fixes. But if you copypaste the necessary code into your folder, you are free to do anything you want with it. This is a big deal in research projects where experimentation is a norm rather than an exception. Especially if you are not sure if changes are going to be useful for everyone. I find that deep learning codebases are suitable for copypasting the most. Typically, the amount of code needed to describe a model and its training is not so huge. But at the same it could be very nuanced and hard to generalize. Shareable training scripts tend to grow to an unmanageable size: e.g. Hugging Face has +4k lines. Interestingly enough, transformers opted for copypaste on the model level. Please check out their with the reasoning behind their "single file model" policy. See more resources about the beauty of copypaste at the end. transformers Trainer post An alternative to copypaste is staying on a branch. But I feel like it brings too much overhead in teamwork. Also, I found several more posts about the beauty of copypaste - see more posts in the conclusion. Maintaining shared research code is hard Maintenance of heavily used shared code requires a lot of work. Take a look at the number of file lines plotted against the version. You can see that even the most advanced research teams struggle to keep the complexity in check. torch.nn.Module Pytorch Don’t underestimate time and resources needed to maintain a large shared research code. The more a research library is used the more complicated it becomes. It happens faster than for a typical library because every research direction has a slightly different usecase. Establish very strict rules of what could be contributed back. Otherwise, the shared code becomes fragile and overgrown with a slew of options, buggy optimizations and edgecases. Since the majority of research code dies out, all this extra complexity will never be used again. Dropping some of your shared code will free up some time to do actual research. Design for exploration, not for code reuse It is somewhat true that you don’t want too much even in production. Try to implement the simplest possible solution that meets the requirements. But in production code there are always aspects of maintainability to consider. For example, error handling, speed, logging, modularization is what you typically need to think about. to future-proof your code In research code, none of that matters. You just want to quickly prove that an idea is good or bad in the fastest possible manner and move on. So the dirty simplicity that achieves it without any modules or APIs is totally ok! Don’t waste valuable time on premature software investments such as: creating component interfaces too early in the project. You'll spend too much time fitting into self-made artificial constraints optimizing deep learning training infrastructure before committing to deep learning solution using production config/factories/serialization systems or base classes. Quite often you don't need their functionality during prototyping overly-strict linting and type checking systems. No reason to slow down fast-changing throw-away research code. 2. Invest into fast exploration The goal of a research project is to find a novel solution. No one knows (by definition) what it looks like. It is akin to an optimization process in a complicated research landscape with limited information. To find a good minimum, you need to try many paths, recognize good and bad paths and not get stuck in local minima. To do all of it fast, you sometimes need to make software investments instead of taking tech debt. Speed up common paths There are several different research paths you want to try. Is there a design, a library or an optimization that would shave off time from the majority of paths? You should be careful to not over-engineer anything because you don’t always know all the ideas you’re about to try. This is very custom to every project, but here are some examples: if you train deep networks, invest into training infrastructure. Figure out hyperparameters allowing you to converge during training quickly and reliably if every experiment requires you to use a different model, figure out how you can quickly swap them (e.g. by using a simple factory system or just copypaste) if every experiment has too many parameters and is hard to manage, invest into a nice configuration library. Branch out quickly Researchers should be able to initiate new diverse ideas quickly. It seems easy at the start of the project. But then it gradually becomes harder and harder as people get entrenched in their favourite research paths. To address this, cultural and organizational changes are essential. There should be a process stopping non-promising research before sinking too much money and emotion into it. Regular demo days and technical peer reviews can serve as effective strategies for this purpose. It's also important to find a balance between people jumping on a new shiny idea and properly closing current projects. But this is a software post, so here are some practices to make branching out new projects easy: keep evaluation code disentangled from algorithms. Evaluations are typically more stable than research directions embrace starting a new project from a blank slate, but then watch out which components are reused. Modularizing and cleaning them up is a good investment in a new research project, implement the most innovative and risky component first. Doing so identifies the majority of bottlenecks guiding future software design. Increase signal to noise ratio Noisy and buggy code makes results so ambiguous and inconclusive that the whole project is going to be a waste of time. While you shouldn't over engineer things, you can easily follow these simple rules of thumb to avoid messy code: avoid code with side-effects default to functions rather than classes; and with classes, prefer encapsulation vs. inheritance minimize length of functions/classes/modules; minimize number of if-statements know python well, but use simple techniques. Resist the temptation going into intellectual weeds of metaclasses, decorators and functional programming. Software that produces different results during different runs is tricky to work with. If you made an important but wrong decision based on an unlucky seed, you are going to waste a lot of time recovering. Here are some tips when dealing with non-deterministic software: understand whether noisy is coming from the algorithm or it's evaluation. Noise sources compound and you should strive for completely deterministic evaluation. don’t stop finding sources of randomness until you really get a reproducible script. Remember that after finding all random seeds, noise might come from data or from generic functions with side-effects. vary seeds and determine baseline variance of your results. Don’t take decisions on non-statistically significant results. Conclusion The punchline comes from about research code: this post . “You don't bother with [good software design] because the code is not the point. The code is a tool that gets you the answer that you need” It is extremely important to have great coding foundations. But at the end of the day, the exploration and actually useful product is what matters. If you use too much production software in research, you waste time needed to discover something new. Instead, find what slows down your exploration process. Speed up research paths by investing in fast branching, time to results and clean noiseless code. It would be crazy to argue completely against code reuse. I just want to point out that code reuse should be a well-balanced activity. In research, the ratio of throw-away code is larger than in production. The balance is tilted further against reuse. Here are few more great posts with pitfalls of code reuse: Reusable Code: The Good, The Bad, and The Ugly When It Makes Sense to Repeat Yourself Balanced StackExchange with pros and cons of reuse And here are few more posts arguing for copypasting practices: Is copypaste really a problem? When Should You Copy-Paste Code? The Good and the Bad of Copy-Paste Programming "As a senior with 5 YOE, I feel like part of my pay is knowing what code to copy lol" Thank you for reading! I feel some bits are a bit controversial, please let me know in the comments! Also appears . here

Software for Exploration: Achieving Balance in Discovery

অনুসন্ধানের জন্য সফ্টওয়্যার: আবিষ্কারে ভারসাম্য অর্জন করা

Software für die Erkundung: Ausgewogenheit bei der Entdeckung erreichen

Software para la exploración: lograr el equilibrio en el descubrimiento

Logiciels d'exploration : parvenir à un équilibre dans la découverte

अन्वेषण के लिए सॉफ्टवेयर: खोज में संतुलन हासिल करना

探索用ソフトウェア: 発見のバランスを実現する

탐험을 위한 소프트웨어: 발견의 균형 달성

Software para Exploração: Alcançando Equilíbrio na Descoberta

Программное обеспечение для геологоразведки: достижение баланса в Discovery

Keşif Yazılımı: Keşifte Dengeyi Sağlamak

Phần mềm khám phá: Đạt được sự cân bằng trong khám phá

用于探索的软件：在发现中实现平衡

The TechBeat: Product Discovery in B2B and B2B2C Environments: A Guide for Product Managers (2/18/2024)

The TechBeat: Product Discovery in B2B and B2B2C Environments: A Guide for Product Managers (2/16/2024)

The TechBeat: Product Discovery in B2B and B2B2C Environments: A Guide for Product Managers (2/14/2024)

How are you, hacker?
🪐 has got you covered with fresh content from our trending stories of the day! Set email preference . Want to know what's trending right now?: The Techbeat by HackerNoon here Product Discovery in B2B and B2B2C Environments: A Guide for Product Managers By [ 5 Min read ]
Product discovery is a key stage in the product development lifecycle. In B2B and B2C businesses, this process has its nuances, which I cover in this article. @dashadobrego Read More. Octopus Network Unveils Omnity, Pioneering a Fully Decentralized Cross-Chain World By [ 2 Min read ]
Octopus Network announces the upcoming launch of Omnity, a groundbreaking product designed to enhance cross-chain transactions @pressreleases Read More. LanceDB: Your Trusted Steed in the Duel Against Data Complexity By [ 8 Min read ]
There are many ways to build on the foundation offered in this tutorial to create performant, scalable and future-proofed ML/AI architectures. @minio Read More. The Principles to Keep In Mind When Building a Modern Datalake for Your AI Infrastructure By [ 5 Min read ]
The AI game is about performance at scale, and this requires the right foundation. Here's how to be smart when building a modern datalake. @minio Read More. What Are Wallet Drainers? How to Protect Yourself From Them! By [ 7 Min read ]
Learn about wallet drainers, how to detect and respond to compromises, and the role of Wallet Rescue by Hackless in safeguarding the crypto assets. @hackless Read More. Understanding Vulnerabilities and Safeguards in Blockchain Oracles By [ 2 Min read ]
This Blockchain Oracle Summit break down offers key insights into the vulnerabilities and safeguards in the realm of blockchain oracles. @oraclesummit Read More. Headset Reality Wars: Meta @ $1B Quarterly VR Revenue & Apple Vision Pro Sells 200k Units in 10 Days By [ 2 Min read ]
Reality Labs, the VR subsidiary of Meta, passed $1B in Q4 VR Revenue, according to Meta’s earning call. Apple Vision Pro sold 200k units in its first ten days. @david Read More. A Step-By-Step Guide For Implementing OKRs In Your Scaling Startup By [ 6 Min read ]
An OKR has two parts: the objective, your goal, and the key results, what success looks like. @vvmrk Read More. Improving Testing Algorithms: Mathematical Approaches in Software Testing By [ 7 Min read ]
Test design, Code coverage, MC/DC, Boolean algebra, Pairwise testing, Combinatorial testing, Orthogonal arrays, System states @shad0wpuppet Read More. Cybersecurity Tips: Vulnerability Scanners Essentials By [ 6 Min read ]
Cybersecurity Vulnerability Scanners Essentials, OWASP ZAP, Burp Suite, Nessus, Sn1per, Metasploit, SQLMap. @shad0wpuppet Read More. Cybersecurity Essentials: Practical Web App Security Testing Tips for QA Engineers By [ 10 Min read ]
Cybersecurity for QA - web app security: XSS, Header Injections, CSRF, RCE, Command Injection, Web Parameter Tampering, CORS, Content Security Policy (CSP) @shad0wpuppet Read More. How to Use AI and PDF Services to Automate Document Summaries By [ 8 Min read ]
My three-page PDF is now one simple paragraph. You can imagine how useful this would be for organizations with millions of documents. @raymondcamden Read More. Why Open Source Language Models Are True “Open AI” By [ 4 Min read ]
H2O.ai's Danube is the latest in a series of open-source language models. @FrederikBussler Read More. Tech Company News Data Dump on HuggingFace: 7M Most Cited Posts About 3k Most Valued Tech Companies By [ 2 Min read ]
HackerNoon curated and open sourced the internet's most cited 7M+ tech company news articles and blog posts about the 3k+ most valuable tech companies. @david Read More. Timeboxing: How I Plan My Day (as a Product Manager) By [ 2 Min read ]
Timeboxing changed the way I plan my days (and even months). I get more of the right 'stuff' done with this method than I ever did before. @alexdebecker Read More. Dione Protocol Shines as Crypto Winter Thaws: Paving the Way for Green Energy Revolution By [ 7 Min read ]
As the cream rises to the top, Dione Protocol is poised to shine, leveraging this revitalized landscape to drive growth and deliver on the promise of blockchain @dioneprotocol Read More. I Built a Platform to Help Users Practice Programming Challenges Guided by AI By [ 6 Min read ]
Walking  through the process of building a platform leveraging AI to help users practice for interviews; integrating feedback from the first few live user tests @emmanuels Read More. The Top 13 Trends in 2024: AI Predictions By [ 12 Min read ]
Read our VIRAL 70k-view Medium article "AI predictions: Top 13 AI trends for 2024". A comprehensive guide to the future of artificial intelligence. @pragmaticcoders Read More. Tech-Gate.org Racks Up Over 1000 DMCA Copyright Infringements By [ 5 Min read ]
Tech-Gate.org, apparent operated and published under the name Bourbiza Mohamed, has been under fire for this rampant copyright infringement site. @technologynews Read More. A Year Since ChatGPT Launch: Do Content Creators Really Need to Fear AI? By [ 6 Min read ]
A closer look at the impact of ChatGPT on writing jobs and the content industry. 🧑‍💻 What happened in your world this week? It's been said that , , . Feeling stuck? We got you covered ⬇️⬇️⬇️ We hope you enjoy this worth of free reading material. Feel free to forward this email to a nerdy friend who'll love you for it.
See you on Planet Internet! With love,
The HackerNoon Team ✌️ @deepikapundora Read More. writing can help consolidate technical knowledge establish credibility and contribute to emerging community standards ANSWER THESE GREATEST INTERVIEW QUESTIONS OF ALL TIME

The TechBeat: Product Discovery in B2B and B2B2C Environments: A Guide for Product Managers (2/12/2024)

DISQO

disqo.com

201-500 employees

Since 2015

COMPANY RANKING

DISQO

EVERGREEN INDEX #3901

Representing this company? Click here to claim and customize this page!

DISQO WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!

DISQO

disqo.com

201-500 employees

Since 2015

COMPANY RANKING

DISQO

EVERGREEN INDEX #3901

RELATED COMPANIESRANK

STORIES ABOUT DISQO

Representing this company? Click here to claim and customize this page!

DISQO WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!