Source: https://internal-agents.com/agents/replit-manager-agent # Replit — Manager agent (agent-of-agents) An internal agent-of-agents stack where every employee gets a manager agent that spawns multiple agents for verifiable work and escalates judgment to humans. - Approach type: Orchestration system - Deployment stage: Scaled - Autonomy: Drafts reviewed - Evidence strength: Detailed primary - Status: Internal - First reported year: 2026 - Work: Coding, Code review, Support, Research, Data - Interfaces: Slack - Invocation: Interactive - Entry reviewed: 2026-09-09 ## Where people stay involved - **objective → verifiable multi-agent work product** — Work product review · Level 3 ## Overview ### Summary An internal agent-of-agents stack where every employee gets a manager agent that spawns multiple agents for verifiable work and escalates judgment to humans. Fact · Reported · High confidence · `replit-manager-agent--summary` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) ## How it works ### Supporting component One human gives an objective; the manager spawns parallel agents for verifiable work and escalates judgment Fact · Reported · High confidence · `replit-manager-agent--primitives-0` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) ## Supervision evidence ### Operating model assessment Level 3 for objective → verifiable multi-agent work product; human attention boundary: work-product-review. Inference · Catalog judgment · High confidence · `replit-manager-agent--operating-models-0` Confidence reason: The source describes autonomous parallel execution followed by human judgment on the resulting work. Qualifications: - Observation date: 2026 Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) ## Implementation details ### Sandbox microVMs and remote filesystems behind access policies, token proxies, audit logging, and a ZeroTrust network Fact · Reported · High confidence · `replit-manager-agent--architecture-sandbox` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) · Preserved content.md, lines 34 ### Harness Fleet/loop orchestration: a manager agent launches parallel agents for verifiable work and escalates judgment Fact · Reported · High confidence · `replit-manager-agent--architecture-harness` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) ### Model Not specified Fact · Reported · High confidence · `replit-manager-agent--architecture-model` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) ### Interfaces slack Fact · Reported · High confidence · `replit-manager-agent--architecture-interfaces` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) ### Tool access Investigates incidents, reviews PRs, answers questions, analyzes company data, triages support, researches sales accounts, improves Replit Agent itself Fact · Reported · High confidence · `replit-manager-agent--architecture-tool-access` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) ### Context management Manager agent coordinates parallel sub-agents and routes results Fact · Reported · High confidence · `replit-manager-agent--architecture-context-mgmt` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) ## Reported metrics ### Headline claim 2.9x code output for a consistent author cohort; review latency, PR reversions, and incident trends reported flat Metric · Reported · High confidence · `replit-manager-agent--headline-metric` Confidence reason: A linked first-party source states the claim. Qualifications: - Reported by: Replit - Scope: Code output for a consistent author cohort across early January to late June; separate from company-wide hiring effects - Denominator: Same cohort of authors before and after - Method: Comparison of contributed code for a consistent author cohort; raw company-wide lines of code increased 5.8x Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) · Preserved content.md, lines 10, 38–56 ### Key observation 2.9x code output for a consistent author cohort from early January to late June Metric · Reported · High confidence · `replit-manager-agent--key-metrics-0` Confidence reason: A linked first-party source states the claim. Qualifications: - Reported by: Replit - Scope: Code output for a consistent author cohort across early January to late June; separate from company-wide hiring effects - Denominator: Same cohort of authors before and after - Method: Comparison of contributed code for a consistent author cohort; raw company-wide lines of code increased 5.8x Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) · Preserved content.md, lines 10, 38–56 ### Key observation No corresponding deterioration in review/reversion/incident metrics Metric · Reported · High confidence · `replit-manager-agent--key-metrics-1` Confidence reason: A linked first-party source states the claim. Qualifications: - The source does not report the denominator of this figure. - Reported by: Replit - Scope: Company code review latency, PR reversion rates, and incidents opened during increased code output - Method: Company comparison of review latency, PR reversion rates, and incident trends Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) · Preserved content.md, lines 50–56 ## Lessons and interpretation ### Lesson Give every employee a manager agent that spawns sub-agents; verifiable work parallelizes, judgment escalates to humans Inference · Catalog judgment · Medium confidence · `replit-manager-agent--lessons-learned-0` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) ### Lesson Track outcome metrics (reverts, incidents), not activity; output can scale without quality regressions Inference · Catalog judgment · Medium confidence · `replit-manager-agent--lessons-learned-1` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [1] [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/replit-manager-agent-source-1/content.md) ## Sources 1. [The Self-Driving Company](https://ld.replit.com/blog/self-driving-company) - Engineering blog · First party · Evidence - Original URL: - Accessed: 2026-08-12 · Last verified: 2026-08-31 - Preserved copy in the repository: