Coding Tools
Fintech scales coding agents with self-serve inference
Together AI Bloghigh confidence
Why it matters
- →Self-service inference allows engineering teams to scale AI workloads instantly without waiting for platform tickets.
- →Prioritizing concurrency over raw throughput is critical for spiky, agent-based coding traffic patterns.
- →Direct API control over model tuning and endpoint configuration reduces operational friction significantly.
A global fintech has deployed Together AI's Dedicated Model Inference to support its internal coding assistant, running the GLM-5.2 model on a cluster of 56 NVIDIA B200 GPUs. The company shifted from a static capacity model to a self-service architecture that prioritizes concurrency headroom over raw throughput, allowing engineering teams to scale endpoints and adjust tuning parameters in real-time. This approach resolved previous issues with request queuing during peak working hours by giving developers direct control over endpoint lifecycle and observability metrics without platform team intervention.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
More in Coding Tools
Coding Toolscoding
Claude Code v2.1.323 adds experimental Code Mode and plugin safety
This release quietly introduces Code Mode, an opt-in feature where Claude writes JavaScript to call its own tools, marking a shift toward more autonomous agent-like behavior within the CLI. The update also hardens the plugin ecosystem by enforcing version ranges and adding strict TypeScript type checking for hooks, which should significantly reduce instability from third-party extensions. Beyond these features, the changelog is dominated by critical fixes for authentication drift in background sessions and memory leaks when handling large files, stabilizing the tool for heavy enterprise use.
Claude Code Releases
Coding Toolscoding
llama.cpp b12336 fixes HunyuanVL crash
This release quietly patches a critical memory overflow in the mtmd module that caused crashes when loading HunyuanVL models. The issue stemmed from F64 position embeddings overflowing an F32 buffer, a subtle bug that broke multimodal inference for users relying on this specific architecture. By enforcing strict F32 requirements, llama.cpp stabilizes support for Tencent's vision-language model without adding new features or changing the core API. It’s a necessary maintenance update that keeps local multimodal workflows from breaking unexpectedly.