The next edition of the Boulder AI Builders meetup is tonight, and currently we're 50% over capacity. It's gonna be fun. 😎 Lots of great content lined up, come hear from local startups Stateless, Quadratic, Knolly, and Returned.com, as well as see some of the latest work NVIDIA is cooking up with their NeMo platform. 🔥 Huge thanks to our long-time sponsor Technical Integrity and our newest sponsor SVB for their help to make these happen. 🙏 We're also announcing the schedule for the next two events in August & September -- including a special edition in Denver for Denver Startup Week! * Wednesday 8/14 in Boulder * Wednesday 9/18 in Denver for DSW RSVP links in the comments! Just a heads up too: Since the RSVP list has grown so big, we're intentionally holding some tickets back for the next ones and will prioritize making sure they go presenters and others in the community who might not get on the list right away. 🙌 Now's your chance to get in early.
Freeplay
Software Development
Boulder, Colorado 608 followers
A better way to build with LLMs. Prompt engineering, testing & evaluation tools for your whole team.
About us
A better way to build with LLMs. Bridge the gap between domain experts & developers. Prompt engineering, testing & evaluation tools for your whole team. Now in private beta.
- Website
-
http://freeplay.ai
External link for Freeplay
- Industry
- Software Development
- Company size
- 2-10 employees
- Headquarters
- Boulder, Colorado
- Type
- Privately Held
- Founded
- 2022
- Specialties
- Artificial Intelligence and Developer Tools
Locations
-
Primary
Get directions
Boulder, Colorado, US
Employees at Freeplay
Updates
-
Our CEO Ian Cairns was invited to contribute to Deloitte Insights for technology leaders. He talks about what's important to know about building with generative AI, and what makes software teams successful. Some of the big ideas, and link in the comments: 📈 Evaluation: "A custom panel of contextually relevant evaluations forms the backbone of analyzing AI products." 🧑💻 Data labeling and curation: "You need people with sufficient domain expertise constantly looking at data. There’s no such thing as full automation when it comes to building great generative AI products." 🧪 Testing: "Testing a generative AI product requires coming up with a representative list of all the possible types of interactions and edge cases that may occur for customers, and making sure each behaves reasonably." 🤷 Why bother? "Generative AI will be a huge competitive advantage for companies, but only if they’re able to make the jump to operate successfully in these new ways. The folks who haven’t yet stepped into that process change often find themselves stuck experimenting and trying to get the confidence they need to even ship to production." What resonates? What did he miss? Tell us in the comments.
-
-
Our latest Freeplay blog post is titled "Building an LLM Eval Suite That Actually Works in Practice." What's the secret? Having a good workflow and process for your team to iterate. So many people are looking for a silver bullet right now for building betterAI products, and if there is one, we think it's being able to iterate quickly. Check out the link in the comments to read more. And if you're curious, Jeremy Silva is giving a talk on the topic next Tuesday at the MLOps Community AIQCon. Also in the comments!
-
-
In SF next week? Join us at AIQCon! Check out Jeremy Silva's talk -- details below and link in the comments. 🙌
Looking forward to next week! We're heading to SF for several AI events including AIQCon hosted by the MLOps Community. Our own Jeremy Silva has been an organizer for the Denver chapter, and I'm excited for him to take the stage representing Freeplay. He'll be talking about "Building a product optimization loop for your LLM features." Interested to check it out? Jeremy's talk will be streaming online at 1 pm PT, or DM me if you're interested to come in person — we've got a few tickets. 🔥 Link with the details is in the comments.
-
-
Human Review + Auto-Evaluations = 🚀 A faster, integrated human review and labeling workflow can be the key to unlocking better AI product performance. We're making it easier to do both: * Human labeling & model-graded evals already work together. Labels get incorporated to the optimization workflow for model-graded evals. * Our new Live Filters feature makes it easier to monitor production data, leading to more labels getting added. The whole system gets stronger as a result. Read more here: https://lnkd.in/gka6Et4p
-
We're talking live with Shane McAllister at MongoDB about the practicalities of building and optimizing RAG systems.
Livestream: Unlocking The Power of Retrieval Augmented Generation
www.linkedin.com
-
Join us tomorrow live on LinkedIn! Our co-founder Ian Cairns and AI engineer Jeremy Silva will be joining our partners at MongoDB for a conversation about the details of testing, evaluating & optimizing RAG systems. Come hear about lessons learned in helping lots of teams build & optimize RAG systems, and see how Freeplay can help.
In this episode, we explore the intricacies of deploying high-performance Retrieval Augmented Generation (RAG) systems in production environments with Ian Cairns, co-founder, and Jeremy Silva, engineer at Freeplay. Learn how Freeplay’s innovative platform, in collaboration with MongoDB, simplifies the complex process of experimenting, testing, and tuning RAG features for large-scale applications. Discover the critical role of effective retrieval strategies, optimal generation models, and continuous system optimization. Whether you’re a developer, product manager, or part of a software team, this episode offers valuable insights into enhancing your LLM applications using MongoDB and Freeplay. Tune in to understand the challenges, solutions, and best practices for achieving reliable and scalable RAG systems. - ✅ MongoDB and Freeplay Overview - https://lnkd.in/e6X_JKXN ✅ Freeplay website - https://freeplay.ai/
The Power of Retrieval Augmented Generation - Freeplay and MongoDB
www.linkedin.com
-
In Austin this Thursday? We're sponsoring the ATX Product Mega Meetup: AI Edition! Come check it out and say hello. 👋
I'll be in Austin this week speaking at the ATX Product Mega Meetup: AI Edition. In town and want to join a couple hundred Product Managers to talk about building AI products? Come join us! 🧠🤠 Freeplay is grateful to be sponsoring and supporting the work done by Women In Product, AI Product Collective, Agile Austin, Product Camp, ATX Product Happy Hour, and The Product League, along with the good folks Pragmatic Institute and Productbot AI. https://lnkd.in/gq6iU2AE
ATX Product MEGA Meetup: AI Edition
eventbrite.com
-
Major product updates recently! Check them out below. 👇
One of our biggest releases yet at Freeplay. 🔥 We've been fortunate to work with a growing list of large enterprises and growth-stage companies building generative AI products at scale. These updates all reflect the needs of production software teams to rapidly evaluate and iterate on AI features with a combination of automation and human review. * Create completely custom evals in your code for both single-answer grading on live data, and pairwise comparisons on test results * Use Freeplay to quickly optimize model-graded evals and confirm their results align to your team's human judgement * Build and save complex filters for easy repetitive reviews of production and test data * Quickly launch head-to-head comparisons between any prompt, model or code change and view the results for a full eval panel — as well as score them for human preference * Plus: Native support for Bedrock Anthropic, Azure OpenAI, and SageMaker Llama 3 makes it easy to use your own privately-hosted models in the Freeplay playground and in-app testing features Here's a quick Loom that shows the alignment workflow for model-graded evals. Link to the full release notes in the comments! Let us know what you think, and feel free to DM with any questions.