Freeplay
@freeplay_ai
Followers
391
Following
257
Media
10
Statuses
146
The end-to-end platform for software companies to ship great products with LLMs. Collaborative tools for evaluating and optimizing production AI products.
Boulder, CO
Joined March 2023
Links to full episode on Spotify, YouTube and Apple Podcasts are here. https://t.co/dz5BP81Sk6
freeplay.ai
Ship better products with LLMs. Freeplay gives product teams the power to prototype faster, test with confidence, and optimize features for customers.
0
0
0
🎙️ Why talk about "AI Employees" instead of agents? Maybe the paradigm of employee feedback and performance management is the future of agent observability and evaluation... On our latest episode of Deployed we talk with @surojit, founder of @Ema_Unlimited, about his lessons
3
0
18
Pleased to share that @freeplay_ai is HIPAA-compliant and our SOC 2 Type II is renewed. We’re supporting Fortune 100 teams with multi-region, private-cloud deployments. Enterprise-grade controls, strong encryption, on-prem options. Details at:
trust.freeplay.ai
Ready to turn trust into your competitive advantage? Sprint through security reviews and quickly share key security information with Trust Center.
0
1
6
🎙️ New Deployed episode with @ryancarson, "Builder in Residence" at @AmpCode. It's a great convo about building code agents and building as a solo founder at the same time, and it's refreshingly honest. Case in point, on evals for coding agents: "We do not have a set of evals
4
11
35
Teams do weeks of prompt engineering around one model. Then another comes out with better performance but nobody wants to re-optimize from scratch. They tell us all the time they stay stuck on an old model because switching feels like too much work. We built automated
freeplay.ai
Ship better products with LLMs. Freeplay gives product teams the power to prototype faster, test with confidence, and optimize features for customers.
0
0
0
The future is using AI to improve AI systems, but the secret is having the right data. Freeplay helps you turn your production data into an automated prompt engineer. Optimize your prompt for any model, using your real logs, evals, customer feedback, and more. Here's how it
1
1
9
We loved having Kelly on Deployed! Check it out.
Do you believe *70% right + fast* is better than *90% right + a bit slower*? Talked with @cairns on the Deployed podcast about this and: - My low-tech approach to initial eval sets - How tech roles are evolving in the AI era - A bunch more tactical advice you can apply in your
0
0
1
Want to know how Google Labs builds AI products like NotebookLM, Jules & Stitch? 🎙️ On the latest Deployed podcast we talked with @kellyschaefer, a Product Director for Labs. It's a great listen for any builder on AI product strategy tactics - like getting started with evals.
1
0
1
New blog post: Onboarding for coding agents I used to write 100s of lines in files like CLAUDE .md Recently, I shrunk it to just 2 small sections: Onboarding & Quality Gates The idea: - Context → READMEs - Constraints → Environment - Start AI sessions with "onboarding"
5
8
41
This is good company to be in. 🙌 If you haven't listened yet, we interview leaders on Deployed who are building AI products at scale and they share their lessons learned. Next guest will be from Google Labs! Check it out. https://t.co/eQ1I17hSau
open.spotify.com
Podcast · Freeplay · Deployed is the podcast for people building AI products.With all the hype about AI over the past two years, it’s often been hard to discern what’s actually working. We started...
I'm enjoying podcasts these days during morning runs. Mostly around Tech+Startups+AI. My favorites: - The Pragmatic Engineer by @GergelyOrosz - Deployed by @freeplay_ai - How I AI by @clairevo - Lenny's podcast by @lennysan - The Startup Ideas @gregisenberg Suggestions?
0
1
6
Highly useful talk for anyone running production AI products! This was a crowded room at @aiDotEngineer Chris Hernandez from Chime and Jeremy Silva from Freeplay share a practical Ops process to continuously improve AI products in production. https://t.co/D1815w4TML
0
1
1
There's a better way to build multimodal AI apps! Check out the latest, our team's been cooking. 👇
Building voice agents or AI image analysis tools? There's a better approach than "vibes-based testing" 🎯 @freeplay_ai help speed up your iteration loop with multimodal evals, observability and experimentation. The video shows how. Here are the key tactics that actually help:
0
1
1
🎙️New Deployed episode with @kwindla! Deep dive on voice agents and voice AI, including his tips for evals and more. Kwin created @pipecat_ai (the most popular open source voice agent framework) and has been building voice AI since before it was cool. Tons of practical advice
1
2
7
Everyone's talking about evals, but far too few AI product teams are actually using them. 😬 @freeplay_ai's playground and testing features make it easy to get started with evals. Set up and run quantitative experiments with your real production prompts in minutes. 📝 Import or
1
1
3
Good summary from our founder @cairns of how Freeplay speeds up AI engineering for production teams. 👇
The best AI teams aren’t just engineering-led anymore. @Freeplay_ai helps domain experts take over evals + QA, cutting cycles from weeks to hours. By putting domain experts in the loop, teams are shortening eval cycles and shipping higher-quality AI features, faster. Full
0
1
1
Just posted our recap of @aiDotEngineer World's Fair 2025. 3000+ builders, all deep into production AI systems. Energy was different this year - much more focused on the hard operational problems of making AI products actually work at scale. It was the best one yet. Two big
2
5
14
This is a great read! And a key part of why we're building @freeplay_ai. PMs use Freeplay daily to run experiments and write their own evals (without needing to wait for engineering or data science). Shoutout to our friend @danielmckinn0n too for the great content here. I'll
When it comes to new skills that PMs have to learn in 2025, evals are at the top of the list. You can't PM without product analytics. You can't PM AI without evals. 1. The people working in AI are saying it loud and clear. - @kevinweil, the CPO of OpenAI - @mikeyk, the CPO
1
2
3
100%. I’ve been talking to a lot of voice ai developers about this, lately. You need basic evals for your conversation flow and tool use, AND you need to monitor in production for latency and api errors. Otherwise you are flying blind. Relatedly, you should not be using preview
I keep saying this - if you rely on an AI model, you need to continuously monitor it like any other critical IT system. Otherwise you're blind to any changes in the models, prompts, or any of the intermediary systems that might affect its behavior.
7
6
93
🎙️ New Deployed episode with @zeddotdev founder @nathansobo is live! Nathan's been building better code editors for 10+ years. Now Zed has some of the most impressive agent AI editing features (including real-time streaming edits). His number one piece of advice: "Automate your
1
4
9