Unify CTO breaks down how the team cut AI agent costs 95% in two weeks
LangChain · x · 2026-08-26
LangChain's Max Agency podcast interviews Unify co-founder and CTO Connor Heggie. Unify builds agents for go-to-market teams — but instead of replacing sales reps, its agent acts as "an engineer in their back pocket."
Key topics: cutting 90-95% of costs two weeks before launch; the 15-requests-per-second ceiling inside OpenAI's prompt cache; why your LLM judge must be a different model family; the Speed Audit (why one at a time beats a table of 1,000); why Unify's subagents are just a function call; and eval lessons Connor carried over from his self-driving startup days and Scale AI.
More from coding & agent
- Build LEGO in the Browser with LLM, LDraw, and Three.js — nptacek · 2026-08-26
- Basecamp goes Agent-Native with CLI and Agent Skill for AI automation — danshipper · 2026-08-26
- Each::labs MCP: Control Premiere Pro via text prompts — aziz4ai · 2026-08-26
- Using Codex/Claude risks giving your business ideas to OpenAI/Anthropic — dbasch · 2026-08-26
- Inth scans your codebase, site and PRs to catch privacy risks before they block deals — paw_lean · 2026-08-26
- Local AI Registry: Open source index for hardware, models, and deployment recipes — StefanoGogioso · 2026-08-26