Suggestions
Murugan Durai
Founder, WiseAnt AI - A GenAI-powered Personal Finance Answer Engine | Engineering & Product Leader
Murugan Durai is a GenAI Consultant at Stealth - Digital Business Services, with a strong background in technology entrepreneurship, software engineering, and product management. He is based in Chennai and has experience with leading dynamic teams and strategic positions in various organizations.
Professional Background
- Current Role: GenAI Consultant at Stealth - Digital Business Services.
- Previous Experience: He has worked with WiseAnt and has been involved in significant projects, including building product knowledge graphs for companies like DoorDash, utilizing large language models (LLMs) such as GPT-4.
Education
- Murugan is an alumnus of Y Combinator, which indicates a solid foundation in startup methodologies and innovation.
Contributions and Insights
He frequently shares insights on LinkedIn about the impact of AI on software sales and the challenges of configuring LLMs, showcasing his expertise in the evolving landscape of technology and artificial intelligence.
For more detailed information, you can view his LinkedIn profile under the username murugandurai.
Highlights
Recently, I came across an article detailing DoorDash's successful utilization of large language models (LLMs), such as GPT-4, to significantly enhance their retail catalog.
https://lnkd.in/gdWjb9ZM
🔥 DoorDash effectively utilized LLMs to automatically extract attributes and implemented retrieval-augmented generation (RAG) to expedite annotation. Furthermore, they explored the potential of multimodal LLMs to process text and images together for attribute extraction.
🎞 Drawing from our experience with building ReelRemy, a movie recommendation system, I am confident that we can successfully apply similar techniques employed by DoorDash.
💡 By leveraging LLMs, we can seamlessly extract relevant entities and relationships from unstructured data sources such as movie plots and reviews. In line with DoorDash's approach, we can also utilize LLMs to discern a user's movie preferences based on their viewing history and ratings.
💡 To further bolster our movie recommendation system, integrating multimodal LLMs to extract information from visual assets like movie posters, trailers, and textual data will undoubtedly enhance the user experience.
💡 Additionally, we could create an LLM-based agent that can search the internet, read reviews, watch trailers, and gather relevant information to provide comprehensive and reasoned movie recommendations tailored to each user's preferences.
#llm #usecase #gpt4 #genai
Do you want to learn about the inner workings of the LLMs, dive here as Anthropic's team dissect their own Claude Sonnet 3 model.
In their own words, this is just the beginning of understanding how LLMs behave with tweaks like when the safeguard is switched off or inputs and they want the LLM incumbents to follow.
It's a long read but worth a try.
https://lnkd.in/gxdX3TEE
PS: I would like to understand how their chat system works specifically with documents like financial results, resumes, graphs, tables, etc
#llm #anthropic #claude3 #genai
