Shayak Sen
@shayaksen
Followers
49
Following
66
Media
2
Statuses
33
CTO @Truera_ai, Solving hallucinations for LLMs with @TruLensML
Joined March 2016
RT @truera_ai: DHS CTO Chris Kraft on Government Applications of GenAI. "In order to measure the “relevancy, accuracy and groundedness” of….
0
1
0
RT @lablabai: The 🌀Gemini AI Hackathon🌀 starts in just 2 days, and we are excited to welcome Shayak Sen (@shayaksen), the CTO & Co-Founder….
0
1
0
Thanks @diginomica and @glawton for covering me and @TruLensML and how systematic evaluation can mitigate hallucinations. #ResponsibleAI.
Trying to build a trustworthy and effective LLM app? Hallucination is your biggest challenge. @diginomica, @glawton deep dives into TruEra's hallucination detection & mitigation workflows. #AIObservability #LLMObservability #LLMapps #AIQuality.
0
0
2
Vanilla RAGs don't really cut it. It takes a lot to get a RAG to work and good evals are at the center of it all. Check this short course with @AndrewYNg @jerryjliu0 and @datta_cs out! And shout back if you'd want to learn more.
Do you want to expand your knowledge of the latest techniques in Retrieval Augmented Generation? . Join our latest course, built in collaboration with @truera_ai and @llama_index, and obtain the tools to build production-ready RAG applications. Join now:
0
2
7
RT @truera_ai: IN 2 WEEKS! . Check out this all-star LLM webinar:. *Moving LLM Apps from Dev to Prod*. Featuring @timt @menloventures, Loga….
0
2
0
* Organizations that are really succeeding at AI have built up the muscle over the older generations of AI .* Data leakage through foundation models is an overhyped risk, instead worry about dataset poisoning (3/3).
0
0
0
* A lot is unknown right now in security for LLMs, anyone who says otherwise is lying.* Observability and security belong together. You can't secure what you can't observe (2/3).
1
0
0
Really glad to had the chance to speak at the Securing AI Summit hosted by @forgepointcap , @BGVcapital and EAIGG about @truera_ai and other topics. We had an impressive lineup speakers that covered the gamut of experiences. Some key hot takes from the sessions: (1/3).
1
0
1
Packed house at @agihouse_org at Build/Break/Test hackathon at 9pm with @llama_index @anyscalecompute @pinecone
0
2
14
RT @datta_cs: Stanford CS 329T: Trustworthy Machine Learning: Large Language Models and Applications -- a course I am teaching with John Mi….
0
48
0
Hallucination testing for RAGs. Testing for these three independent criteria connects the correctness of your answer to the correctness of your knowledge base. Check out this blog by @LofredM
0
1
5
Curious to try a different LLM hackathon? Check out the Build/Break/Test hackathon at @agihouse_org. Sign up at:
AGI House Build/Break/Test LLM Apps Hackathon with TruEra/TruLens and partners @llama_index & @pinecone on Oct 7, 12-10pm. @jerryjliu0 @shayaksen @_jreini @Yuvnesh . Join us for this exciting and unconventional hackathon with key players in the emerging LLMOps tech stack!.
0
1
2
Excited to be speaking at @aicampai Dev Day at @github SF. Join us to learn more about hallucinations and agents.
TOMORROW - Join us in SF for the 5th AI Dev Day!. @shayaksen shows you how to evaluate LLM apps and minimize hallucination. Come hang out and talk #GenAI. @aicampai #AIDevDay #LLMapps #LLMtesting #LLMtracking.
0
0
1
RT @TruLensML: In user-facing LLM apps, #streaming represents a huge gain for #ux. But this can be a tough problem for #eval. You want to….
0
1
0
LlamaIndex let you build really powerful agents really quickly. Next step: evaluation. Thanks @jerryjliu0 !.
Properly evaluating agents is an under-explored problem - what are the key tests you should run on inputs/outputs? 🧪🤖. Check out this brand-new notebook that helps to explore that on our @llama_index Yelp agent! 🧑🍳. Huge s/o to the @truera_ai team:
0
2
12
RT @TruLensML: 🚨 Release Radar - TruLens 0.5.0 is out! 🚨. We upgraded the record viewer 👀, added a simple CLI (s/o @aashsach), support for….
0
2
0
RT @TruLensML: Catch @truera_ai CTO and @trulensML leader @shayaksen at the .@pinecone and @microsoft AI Transformation Summit tomorrow! S….
0
3
0
RT @TruLensML: Evaluate and Monitor the Experiments With Your LLM App from Rink Konstantin via @TDataScience. #fee….
0
2
0
RT @TruLensML: TruLens 0.4.0 now available!. Backwards compatibility for previous versions, a ton of new examples, & a community update (sh….
0
4
0
RT @llama_index: Proper evals are important when setting up a complex retrieval-augmented system:.- Overall App I/O.- LLM I/O.- Retrieved c….
0
13
0