AI Arxiv

Improving search relevance in hyperlocal food delivery using (small) language models References: - https://bytes.swiggy.com/improving-search-relevance-in-hyperlocal-food-delivery-using-small-language-models-ecda2acc24e6

Swiggy’s Generative AI Journey: A Peek Into the Future References: - https://bytes.swiggy.com/swiggys-generative-ai-journey-a-peek-into-the-future-2193c7166d9a

How Salomatic Uses DSPy & Langtrace to Build a Reliable Medical Report Generation System References: - https://langtrace.ai/blog/case-study-how-salomatic-used-langtrace-to-build-a-reliable-medical-report-generation-system

Speak.com's Live Roleplays powered by OpenAI Realtime API References: - https://www.speak.com/blog/live-roleplays

How Meta animates images at scale References: - https://engineering.fb.com/2024/08/14/production-engineering/how-meta-animates-ai-generated-images-at-scale/

Hermes: A Text-to-SQL solution at Swiggy References: - https://bytes.swiggy.com/hermes-a-text-to-sql-solution-at-swiggy-81573fb4fb6e

How Meta trains large language models at scale References: - https://engineering.fb.com/2024/06/12/data-infrastructure/training-large-language-models-at-scale-meta/

The secret ingredient in Continue's autocomplete prompt. References: - https://blog.continue.dev/root-path-context-the-secret-ingredient-in-continues-autocomplete-prompt/

Dropbox - Bringing AI-powered answers and summaries to file previews on the web. References: - https://dropbox.tech/machine-learning/bringing-ai-powered-answers-and-summaries-to-file-previews-on-the-web

ColPali: Efficient Document Retrieval with Vision Language Models Reference: - https://blog.vespa.ai/the-rise-of-vision-driven-document-retrieval-for-rag/

Deep dive on how Headstart, a software development company, utilizes Claude, a large language model developed by Anthropic, to dramatically accelerate software development processes. References: - https://www.anthropic.com/customers/headstart

DoorDash's path to high-quality LLM-based Dasher support automation References: - https://careers.doordash.com/blog/large-language-modules-based-dasher-support-automation/

Emerging Architectures for LLM Applications References: - https://a16z.com/emerging-architectures-for-llm-applications/

How companies are bringing AI applications to life References: - https://www.sequoiacap.com/article/llm-stack-perspective/

Scaling Productivity with Ava — Instacart’s Internal AI Assistant References: - https://www.instacart.com/company/how-its-made/scaling-productivity-with-ava-instacarts-internal-ai-assistant/

How Graphite uses GPT to optimize code reviews References: - https://graphite.dev/blog/how-we-use-openai-to-optimize-code-review

How Salesforce built XGen, their LLM References: - https://engineering.salesforce.com/developing-the-new-xgen-salesforces-foundational-large-language-models/

How Salesforce's Einstein 1 Platform Manages Massive Data and AI Workloads at Scale References: - https://engineering.salesforce.com/how-the-new-einstein-1-platform-manages-massive-data-and-ai-workloads-at-scale/

How Salesforce operationalizes Models at scale References: - https://engineering.salesforce.com/how-the-einstein-team-operationalizes-ai-models-at-lightning-speed-and-massive-scale/

How Pinterest built text-to-sql References: - https://medium.com/pinterest-engineering/how-we-built-text-to-sql-at-pinterest-30bad30dabff

Lessons from Uber’s Tech-Wide Hackathon References: - https://www.uber.com/en-SE/blog/the-transformative-power-of-generative-ai/

How Meta uses LLMs for efficient incident response References: - https://engineering.fb.com/2024/06/24/data-infrastructure/leveraging-ai-for-efficient-incident-response/

How Salesforce Supercharges AI: The Secret to Ensuring Model Reliability and Performance References: - https://engineering.salesforce.com/how-salesforce-supercharges-ai-the-secret-to-ensuring-model-reliability-and-performance/

How AI and Automation Slash Thousands of Manual Hours Annually References: - https://engineering.salesforce.com/aiops-engineering-secrets-revealed-how-ai-and-automation-slash-thousands-of-manual-hours-annually/

LLark: A Multimodal Foundation Model for Music References: - https://research.atspotify.com/2023/10/llark-a-multimodal-foundation-model-for-music/

Stripe leverages GPT-4 to streamline user experience and combat fraud. References: - https://openai.com/index/stripe/

Deep dive into how Canava uses LLMs to generate incident summaries. References: - https://www.canva.dev/blog/engineering/summarise-post-incident-reviews-with-gpt4/

Uber's QueryGPT uses large language models (LLM), vector databases, and similarity search to generate complex queries from English questions that are provided by the user as input. This deep dive chronicles our development journey over the past year and where we are today with this vision. References: - https://www.uber.com/blog/query-gpt

Deep dive and learn how Replit runs LLMs in production References: - https://blog.replit.com/llms

Deep dive into how snowflake built the most powerful text-to-sql model References: - https://medium.com/snowflake/inside-snowflake-building-the-most-powerful-sql-llm-in-the-world-95114114aab9

Deep dive into how a popular VC firm is building GenAI tools with Databricks. References: - https://www.databricks.com/blog/how-vc-firm-building-genai?utm_source=twitter&utm_medium=organic-social

How Zoom built Zoom AI with their federated approach of using Meta's Llama References: - https://ai.meta.com/blog/llama-zoom-ai-companion/https://ai.meta.com/blog/llama-zoom-ai-companion/

Deep dive into how Notion built Notion AI References: - https://www.youtube.com/watch?v=LEIfoMjDDXs

Deep dive into how Shopify improved it's LLM chatbot's streaming experience. References: - https://shopify.engineering/sidekicks-improved-streaming

Deepdive into how Google grounds LLMs in real world or contextual information by coupling two novel techniques - RIG and RAG. References: - https://research.google/blog/grounding-ai-in-reality-with-a-little-help-from-data-commons/

Deep dive into how Uber built a GenAI gateway. References: - https://www.uber.com/blog/genai-gateway/

In this deep dive, learn how Curosor, an AI powered code editor built a near instant AI code editing feature using a unique technique called speculative decoding. References: - ⁠https://www.cursor.com/blog/instant-apply⁠ - https://fireworks.ai/blog/cursor

Deepdive into how Ramp automated receipt processing using LLMs References: - https://modal.com/blog/ramp-case-study

Deep dive into how Replit uses LLMs to understand code at scale. References: - https://blog.replit.com/how-replit-makes-sense-of-code-at-scale-ai-data

In this episode, we analyze AI Engineer job openings at some of the top tech companies in the Bay Area and deep dive into what sort of common requirements and experience these companies are looking for References: - Open job postings on LinkedIn

A quick deep-dive into how Asana tests and evaluates LLMs. References: - https://asana.com/inside-asana/llm-testing-claude-sonnet-analysis

CB-GPT - The opportunity and the vision for GenAI at Coinbase. References: - https://www.coinbase.com/blog/cb-gpt-the-opportunity-and-the-vision-for-genAI-at-coinbase

Video annotator: a framework for efficiently building video classifiers using vision-language models and active learning References: - https://netflixtechblog.com/video-annotator-building-video-classifiers-using-vision-language-models-and-active-learning-8ebdda0b2db4

Let's deep down into how Box.com leverages Amazon Q to power Box AI, their new Generative AI powered offering. References: - https://aws.amazon.com/blogs/machine-learning/discover-insights-from-box-with-the-amazon-q-box-connector/

Wendy’s is Introducing Industry-Leading Technology to the Restaurant Experience. References: - https://www.wendys.com/blog/how-wendys-using-ai-restaurant-innovation

DragonCrawl: Generative AI for High-Quality Mobile Testing. References: - https://www.uber.com/blog/generative-ai-for-high-quality-mobile-testing/

Ava is an internal AI assistant powered by OpenAI’s GPT-4 model. Ava comes equipped with an array of features including conversation search, automatic model upgrades, and a unique “Prompt Exchange” that encourages users to share and reuse prompts, enhancing collaborative productivity. References: - https://tech.instacart.com/unlocking-efficiency-how-ava-became-our-ai-productivity-partner-f1a560686361

LinkedIn has been working hard to develop a new AI-powered experience. They have reimagined how their members go about their job searches and browse professional content. References: - https://www.linkedin.com/blog/engineering/generative-ai/musings-on-building-a-generative-ai-product

In this deep dive, we will look at how Uber is accelerating it's Generative AI efforts internally to solve various engineering problems. References: - https://www.uber.com/en-SG/blog/from-predictive-to-generative-ai/

Originally published in the official blog of DoorDash, this episode deep dives into how DoorDash uses LLMs for building product knowledge graphs. References: - https://careers.doordash.com/blog/building-doordashs-product-knowledge-graph-with-large-language-models/