April 2023 Summaries

OpenAI has released ChatGPT, a large language model that significantly improves complex language generation, revolutionizing the use of generative AI technology in companies. However, building powerful AI-powered applications while controlling costs and engineering prompts is a new challenge for engineering teams. Superblocks now offers a deep generative AI integration with OpenAI, providing an easy layer on top of every OpenAI API and model to ship any AI-powered internal tooling experience in minutes. This includes access to GPT-4, Whisper model, image generation, and more through a user-friendly interface. Superblocks also simplifies prompt engineering by allowing users to reference data using variables in code right within the Prompt and System Instruction fields. Cost control can be implemented by setting token limits for each request, effectively controlling maximum spend and avoiding unexpected overages. The platform integrates with various data sources such as Salesforce, Marketo, Zendesk, and more, allowing users to create business apps for personalized outbound campaigns and streamline customer support processes. Superblocks also supports code generation, text and code summarization, and audio transcription using OpenAI's powerful Whisper API. Additionally, the platform offers granular role-based access control (RBAC), audit logs, observability, single sign-on, system for cross-domain identity management (SCIM), and an on-premise agent to protect sensitive data.

Cortex, a Developer Portal, aims to foster reliability and ownership among engineering teams while allowing them to focus on building software for businesses. As an early-stage SaaS company, Cortex focuses on two main areas: building core product and customer acquisition and growth. To maintain high velocity in these areas, staff should concentrate on their respective expertise. Cortex faced challenges when creating internal tooling for managing customers, software licenses, and mailing lists. They required flexible applications that could provide a single source of information across multiple data sources while adhering to security requirements due to handling sensitive customer data. Previously using Retool, Cortex experienced low adoption rates and difficulty in building out applications. Cortex used Superblocks to build an internal customer application in just hours, enabling their Customer Success team to better serve clients while allowing developers to focus on core product features. By connecting into existing databases and CRM systems, they built an application that provided insight into their customers' account statuses, domains, and observability metrics. Security measures were implemented through deep code-level RBAC, ensuring only select individuals had access to customer data. Cortex also used Superblocks to build a software license generation application for on-premise customers, addressing revenue implications by tracking usage and creating JWT licenses that capped the number of users. The increased efficiency gained from building applications within Superblocks allowed engineers at Cortex to return to revenue-generating work. The team plans to automate more processes using scheduled jobs and workflows, reducing engineering bottlenecks and human error. They are available to answer questions and provide custom demos showcasing how Superblocks can fit specific needs.