I'm using deepseek V4 on two rtx 6000 pros and its working great. Opus is so slow that I get deepseek to do most of the work and Opus is only used to validate and help plan.
Yeah, I really don’t see how this is a sensible interview question. It does not even mention async await syntax. Expecting knowledge on callbacks seems dated.
True, the biggest bottleneck is formulating the right task list and ensuring the LLM is directed to find the relevant context it needs. I feel LLMs in their instruction following are often to eager to output rather than using tools (read files) in their reasoning step.
We’re building conversational product discovery tools for e-commerce stores, moving beyond the limitations of traditional search bars. Our system lets shoppers explore and find products naturally—using their own words. We’re about to launch with our second customer, and early results show a faster, more intuitive, and more convenient shopping experience. For our retail customer we've had users just copy and paste their complete shopping list and be done within one conversation turn. https://www.isartech.io/
Great summary of the trade-offs in Agentic systems. We’ve tackled these exact challenges as we built out our conversational product discovery product for e-commerce at IsarTech [0].
I agree function composition and structured data are essential for keeping complexity in check. In our experience, well-defined structured outputs are the real scalability lever in tool calling. Typed schemas keep both cognitive load and system complexity manageable. We rely on deterministic behavior wherever possible, and reserve LLM processing for cases where schema-less data or ambiguity is involved. Its a great tool for mapping fuzzy user requests to a more structured deterministic system.
That said, finding the right balance between taking complexity out of high entropy input or introducing complexity through chained tool calling is a tradeoff and balance that needs to be struck carefully. In real-world commerce settings, you rarely get away with just one approach. Structured outputs are great until you hit ambiguous intents—then things get messy and you need fallback strategies.