Jefferies built a trading assistant with Claude, Strands, Bedrock and MCP
AWS ML Blog · rss · 2026-07-24
AWS describes how Jefferies built an AI trade assistant for front-office equity traders using Strands Agents, Amazon Bedrock, Claude, Bedrock Knowledge Bases, and MCP tools.
The assistant lets traders ask natural-language questions over trade repositories, FIX messages, and in-memory databases, then generates SQL, executes queries, and surfaces results inside the trading UI. AWS says the system adds guardrails such as content moderation, PII filtering, row-level entitlements, and conversation logging. The post also argues that Strands plus MCP make the architecture extensible, maintainable, and easier to test as new data sources and tools are added.
More from coding & agent
- Point-free code may look better once machines write and verify it — satnam6502 · 2026-07-24
- Use a second model to approve tool calls and review an agent’s trajectory — corbtt · 2026-07-24
- PRO-LONG gives LLM agents searchable programmatic memory and lifts ARC-AGI-3 by 18 points — dair_ai · 2026-07-24
- A CLI idea to capture coding-assistant prompts as git notes — rseroter · 2026-07-24
- Codex bypassed a hidden anti-bot honeypot while a human got blocked — corbtt · 2026-07-24
- Kimi K3 tops Frontend Code Arena and ties the best model in a designer blind test — yuwen_lu_ · 2026-07-24