Emily Ekdahl @emekdahl

AI/LLM Ops Engineer linkedin.com/in/emilyekdahl/ Chicago, IL Joined October 2012

Tweets

534
Followers

210
Following

330
Likes

1K

Emily Ekdahl @emekdahl

2 weeks ago

Can #GPT5 actually do taxes? We ran it on @ColumnTax’s TaxCalcBench. Full return: 30.4% strict ✅ | 53.4% lenient 🤔 Line items: 80.6% strict | 85.4% lenient 📊 Line accuracy is strong. Whole-return accuracy? Not IRS-ready yet. github.com/column-tax/tax… #taxCalcBench #AI #tax

0 0 3 113 0

Hamel Husain @HamelHusain

3 months ago

The most useful bit of my system prompt is this If I provide any feedback on how to improve something, suggest improvements to my prompt that I can make to avoid similar mistakes in the future. Put any prompt improvement suggestions in separate <prompt-improvement> tags.

7 12 274 27K 391

Emily Ekdahl @emekdahl

3 months ago

Can't say enough good things about the AI evals course run by @sh_reya and @HamelHusain! It is informed by real production work across dozens of clients. The opportunities and challenges resonate with my experience evaluating & deploying production AI products.

0 1 7 1K 2

Leonie @helloiamleonie

a year ago

2023 vs. 2024 2023: Vector search is all you need 2024: Evaluate vector/hybrid search against BM25 baseline 2023: „Look, this prompt works!“ 2024: Prompt optimization with DSPy 2023: … 2024: Evals with AI-as-a-judge We‘ve come a long way, but we’re still so early.

0 17 113 6K 45

François Chollet @fchollet

a year ago

Getting employees to work hard and deliver really isn't a matter of mandating work-from-office and long hours. It's a matter of incentives and ownership. People do their best when they work on interesting problems, in a self-directed manner, and get rewarded for success. This…

74 181 2K 87K 337

Adam Grant @AdamMGrant

a year ago

Insecure leaders ridicule others. Secure leaders laugh at themselves. The ability to make fun of yourself opens the door to candor. It’s a mark of humility and a catalyst for learning. Great leaders take their work seriously, but they don't take themselves too seriously.

30 624 3K 136K 310

Emily Ekdahl @emekdahl

a year ago

I worked this demo and found it super helpful in understanding multi-agent architecture. Especially liked the concierge and the continuation agent; I thought they made the experience more fluid for the end user. Thanks, @seldo !

Jerry Liu @jerryjliu0

a year ago

6 74 474 53K 565

Download Image

2 0 2 156 0

Leonie @helloiamleonie

a year ago

What‘s the difference between similarity search and re-ranking? Both return a list of sorted items. Similarity search: • retrieves a specified amount of objects from the entire database • based on their proximity in vector space • simple and light-weight algorithm • fast at…

0 19 94 7K 68

Emily Ekdahl @emekdahl

a year ago

One of the most interesting advances in explainability I’ve seen lately!

elvis @omarsar0

a year ago

One of the most interesting advances in explainability I’ve seen lately!

16 469 2K 121K 3K

Download Video

0 0 1 107 0

Emily Ekdahl @emekdahl

a year ago

It is both exhilarating and humbling to see the pace of change in AI and how quickly hard problems become solved problems!

LlamaIndex 🦙 @llama_index

a year ago

It is both exhilarating and humbling to see the pace of change in AI and how quickly hard problems become solved problems!

6 97 432 71K 681

Download Image

0 0 0 59 0

Adam Grant @AdamMGrant

a year ago

Simone Biles taught a valuable lesson at the Olympics: Overcoming adversity takes time. Resilience is not about how quickly you bounce back—it’s about how fully you recover. You can’t judge people’s strength when they fall. Their fortitude is visible when they rise again.

74 2K 10K 548K 954

Emily Ekdahl @emekdahl

a year ago

Great summary of lessons learned from #AI in production

Eugene Yan @eugeneyan

a year ago

Great summary of lessons learned from #AI in production

19 164 755 89K 964

0 0 2 98 0

Andrew Ng @AndrewYNg

a year ago

Multi-agent collaboration has emerged as a key AI agentic design pattern. Given a complex task like writing software, a multi-agent approach would break down the task into subtasks to be executed by different roles -- such as a software engineer, product manager, designer, QA…

89 517 2K 413K 2K

Emily Ekdahl @emekdahl

2 years ago

The ratio of information gems / minute during @jerryjliu0 's @mlopscommunity talk about Production RAG Pain Points and Solutions was exhilarating! What a great talk!

0 0 2 62 0

Andrew Ng @AndrewYNg

2 years ago

New short course on Building Applications with Vector Databases, taught by @pinecone’s @timt! At the heart of a vector database is the ability to store a collection of vectors and then query against that, meaning input a new vector and find similar ones. This is useful for many…

65 205 1K 137K 652

Download Video

Yann LeCun @ylecun

2 years ago

How enterprises are using open source LLMs: 16 examples. Many use Llama-2: Brave, Wells Fargo, IBM, The Grammy Awards, Perplexity, Shopify, LyRise, Niantic.... Quote: “A lot of customer are asking themselves: Wait a second, why am I paying for super large model that knows very…

49 283 2K 381K 1K

Adam Grant @AdamMGrant

2 years ago

In times of tragedy, don't mistake inaction for apathy. It's often the result of empathy. Empathic distress is feeling helpless in the face of suffering. It drains us until we're numb. You never know whose pain others are carrying. They may need compassion more than you…

53 853 3K 277K 495

Download Image

LlamaIndex 🦙 @llama_index

2 years ago

Today we’re launching a repo that lets you setup a production ETL pipeline for your RAG/LLM app 💫 Index thousands of documents in seconds ⚡️ (and orders of magnitude faster than running on your laptop). It’s a full architecture which bundles LlamaIndex with other popular…