AI Engineering @Redisinc | trying to figure out what’s up | UVA alum 2x
Richmond, VA
Joined November 2012
- Tweets2.1K
- Following795
- Followers467
- Likes7.2K
❤️Redis Iris is the culmination of years of work on building memory-efficient & scalable retrieval systems for recsys, feature stores, RAG and now agents.
Hats off to the entire product, RnD, marketing, DevRel, and leadership team for pulling off this release just in time!
Announcing Redis Iris. The context layer agents have been missing. Redis is already the most used database for agent data
Today we’re making it easier than ever with Iris.
redis.io/iris/
Soooo much work has gone into refining our understanding of how agents best navigate and fetch context. MCP is really just the beginning of course. As a platform, it improves with usage thanks to the agent memory layer and best-in-class semantic cache (FAQs and Procedural memory)
My current view is that legacy data infrastructure is largely NOT ready for the demand that agents will bring. That’s why we are investing in this product area for @Redisinc 🙏
Tyler Hutcherson retweeted
Announcing Redis Iris. The context layer agents have been missing. Redis is already the most used database for agent data
Today we’re making it easier than ever with Iris.
redis.io/iris/
Tyler Hutcherson retweeted
Cool to see how OpenAI uses Redis to scale voice AI to 900M users:
openai.com/index/delivering-…
We're huge Codex fans here at Redis - I use Codex 5.5 extensively - amazing work by the team there. GPT 5.5 is the best model for most tasks IMO. It even fixed my wifi tonight (no joke)
Tyler Hutcherson retweeted
We have been helping developers build out their agent infrastructures, launch their first AI apps and we’ve been building a collection of tools and products based on what we have learned.
Check them out. Give us your feedback. redis.io/ai-incubator/
Tyler Hutcherson retweeted
Stephen Miller: he was a domestic terrorist
The reality: he was an ICU nurse who was trying to help a woman
ChatGPT may have just single-handedly saved Thanksgiving today. Don’t let my wife’s family know 😏
#secretingredient #macncheese #chatgpt #thanksgiving
One of the joys over the last few months has been collaborating on this new @DeepLearningAI course on semantic caching in agentic AI systems. Huge thanks to the whole @Redisinc team (@ilzhechev) and @AndrewYNg for the partnership and support! 🙏🏼
New course announcement: Semantic Caching for AI Agents, taught by @tchutch94 and @ilzhechev from @Redisinc.
Semantic caching can significantly reduce your AI application's inference costs and latency. If someone asks "How do I get a refund?" and another later asks "I want my money back," semantic caching recognizes these mean the same thing so it can use a cached response instead of making another model call.
This short course takes you from building your first semantic cache from scratch to implementing production-ready systems using Redis' open-source tools.
Skills you'll gain:
- Build semantic caches from scratch, then implement them using Redis' SDK with production features
- Measure cache performance using hit rate, precision, recall, and latency
- Enhance accuracy with threshold tuning, cross-encoders, LLM validation, and fuzzy matching
Join and learn to reduce your agentic AI's costs and improve speed! deeplearning.ai/short-course…
Still hiring for a high impact applied AI engineer at @Redisinc in the Bay Area. Looking for a role with high agency and influence? shoot me a DM:
- you understand how to build and scale agentic systems
- you obsess over customer pain
- you seek and build repeatable solutions
Applied AI Engineering at Redis goes way back to when me @SamPartee and @taimurrashid began the team to incubate the AI business for Redis.
Many years later, we’ve grown in impact and visibility even more!
Think of it as a “forward deployed engineering team”
Tyler Hutcherson retweeted
If you can't understand how good your product is, you don't have a product in the first place.
Everyone needs evals (even if they don't understand what "evals" actually mean).
The number one problem I've had to fix when working with teams that can't make progress because their product is running in circles:
1. Find out the metric we care about the most
2. Automate the evaluation of the product
3. Error analysis to focus on the most critical issues and fix them.
I really liked @sh_reya's post, and I love the framing of the question teams should ask:
"When can we afford to be less rigorous, and when can we not?"
Tyler Hutcherson retweeted
🎉10K⭐️🎉
Go check out the repo for tutorials, components, and more to help you build agents.
Incredible milestone!
My Agents-Towards-Production GitHub repository just crossed 10,000 stars in only two months! 🌟
github.com/NirDiamant/agents…
Here's what's inside:
✅ 33 detailed tutorials on building the components needed for production-level agents.
✅ Tutorials organized by category.
✅ Clear, high-quality explanations with diagrams and step-by-step code implementations.
✅ New tutorials are added regularly.
✅ I'll keep sharing updates about these tutorials here.
A huge thank you to all contributors who made this possible!
♻️ Repost to share with others
Tyler Hutcherson retweeted
Join @Redisinc and @langchain on July 23 for a webinar to learn how LangGraph + Redis make it easy to build AI agents with real memory. In the live webinar, you'll get demos, performance tips, and hear from the engineers behind the integration.
👉 RSVP here: events.redis.io/breakingdown…
Tyler Hutcherson retweeted
The fastest serving engine for LLMs is here (open-source)!
LMCache is an LLM serving engine designed to reduce time-to-first-token and increase throughput, especially under long-context scenarios.
It boosts vLLM with 7x faster access to 100x more KV caches.
100% open-source!
Tyler Hutcherson retweeted
🔥 A FREE goldmine of tutorials for building production-ready AI agents! 🔥
Just launched a GitHub repo packed with 25 hands-on tutorials that walk you through every core component of an AI agent pipeline.
🔗 Check it out here: github.com/NirDiamant/agents…
The tutorials are grouped into:
1. Orchestration
2. Tool integration
3. Observability
4. Deployment
5. Memory
6. UI & Frontend
7. Agent Frameworks
8. Model Customization
9. Multi-agent Coordination
10. Security
11. Evaluation
Found it helpful? Drop a ⭐ on the repo!
Tyler Hutcherson retweeted
ok. enough is enough.
RAG vs agentic retrieval.
you 👏 are 👏doing 👏 retrieval. Call grep? Retrieval. Do a vector search? Retrieval. Use BM25? Retrieval.
This dichotomy is just marketing by people with an incentive to make you use their product, for fear of not keeping up with the latest thing.
These are all valid techniques to search data, and give it to a model. The marketing is a distraction. Reason about and measure what works for your application. Decide what you do based on that.