What are some ideas we can potentially execute in tier 2 cities. Most of the innovations are saturated around metro cities.
I think non metros posses great potentials as well. Comments appreciated :)
I feel the scope in ASR has drastically increased especially in the human engaged call centres where system understands the context and helps in capturing user responses better.
And yeah...more data to work with :)
JSON Prompting is so underrated
Want LLMs to respond in clean, structured formats? Here we go...
JSON prompting is the secret sauce.
- Predictable output
- Easy to parse
- Great for APIs & function calling
*Prompt smart. Parse easy.*
#AI#PromptEngineering
Some real world use cases could be:
•Chatbot logs for CRM
•Structured task extraction
•Generating config files
•Extracting entities for search indexing
•Passing outputs to downstream APIs or pipelines
Whenever you suceed building a PoC on a model, some billion dollar company releases a new ASR inference with better efficiency, accuracy and efficacy 😭...Why ?
I've been working with Language models from past year and following the trend.
Surprisingly, LLM Evaluation is turning out to be the next big trend in the AI space.
I'll be learning more on LLM evaluation libraries (RAGAS, Deepcheck etc).
Suggestions and tips appreciated 😁