Uploaded September 2026 | Updated September 2026, 2 weeks ago
Your coding agent isn't burning your budget in the planning step — it's burning it in the executor, where dozens of edits, test-fix cycles, and parallel candidate solutions rack up inference calls fast.
Join Kwasi Ankomah, Lead AI Architect at SambaNova Systems, for Part 2 of the SambaNova Sponsored Webinar Series, where he breaks down how parallel execution, best-of-N selection, and disaggregated serving turn unpredictable agent costs into a controllable line item. Kwasi brings 15 years across financial services, consulting, government, and tech startups to a conversation that goes deep on the engineering choices behind agentic AI in production.
You'll learn:
→ Why the executor — not the planner — is where coding agents actually spend time and money
→ How fanning work across parallel executors changes the economics of running more candidate solutions
→ What best-of-N selection is, and how test suites pick winning candidates from parallel attempts
→ Why fast, affordable inference at scale is a prerequisite for test-time compute to pay off
→ How SambaNova's disaggregated serving architecture affects throughput, latency, and utilization tradeoffs
→ How to evaluate your own agent infrastructure for where executor costs are really coming from
Register now to save your spot.
Your coding agent isn't burning your budget in the planning step — it's burning it in the executor, where dozens of edits, test-fix cycles, and parallel candidate solutions rack up inference calls fast.
Join Kwasi Ankomah, Lead AI Architect at SambaNova Systems, for Part 2 of the SambaNova Sponsored Webinar Series, where he breaks down how parallel execution, best-of-N selection, and disaggregated serving turn unpredictable agent costs into a controllable line item. Kwasi brings 15 years across financial services, consulting, government, and tech startups to a conversation that goes deep on the engineering choices behind agentic AI in production.
You'll learn:
→ Why the executor — not the planner — is where coding agents actually spend time and money
→ How fanning work across parallel executors changes the economics of running more candidate solutions
→ What best-of-N selection is, and how test suites pick winning candidates from parallel attempts
→ Why fast, affordable inference at scale is a prerequisite for test-time compute to pay off
→ How SambaNova's disaggregated serving architecture affects throughput, latency, and utilization tradeoffs
→ How to evaluate your own agent infrastructure for where executor costs are really coming from
Register now to save your spot.
