Dec 19, 20234 min read
How To Fix OpenAI Rate Limits & Timeout Errors
LLMs are starting to get used across a wide variety of use cases. These include translation, sentiment analysis, generating code, blogs…
aiplatformreliability
Read storyPuneet Bhatt
Engineering @ Thine
I build features that make AI feel effortless to use. At Merlin I focus on shipping new ways for people to collaborate with LLMs inside the tools they already know.
- Engineering @ Merlin AI / Thine, shaping AI so people can collaborate with LLMs where they already work.
- Co-built Intwiff, a bike rental SaaS with 200+ active riders and vendor tooling for inventory, fulfillment, and payments.
- Comfortable across React, TypeScript, AWS, Django, LLMs, agents, context engineering.
Blog
Notes from the journey
Essays on building Merlin, staying reliable with AI tooling, and the product experiments that don’t make the highlight reel.
Drafting longer-form stories on product craft, AI workflows, and the business of shipping. Drop topics you want covered via email.
Browse all postsStay in the loop
Ship notes straight to your inbox
I send lightweight updates when fresh essays go live—no spam, just field notes from building products and wrestling with AI workflows.
Prefer RSS? Follow here.