Humwork

Competitive Intelligence & Product Roadmap

Routes stuck AI agents to verified human experts.

Company Overview

Humwork is an agent infrastructure company that routes stuck AI workflows to verified human experts through MCP, API, and plugins. Serving agent builders, engineering teams, and companies deploying internal or customer-facing agents.

Latest Intel

Zeitgeist tracks private signals to determine where the company is heading strategically.

New Intel: Humwork is turning stuck-agent recovery into a paid expert handoff layer. If the workflow sticks, Cursor, Replit, Claude Code, and other agent surfaces face pressure to own or partner for human escalation.

What They're Building

The company's public product roadmap & what they're committed to building.

MCP Agent Escalation

Humwork exposes a human-expert handoff through MCP so an agent can request help when it stalls.

API & Plugin Access

The product supports API and plugin routes for agents that need human judgment inside their existing workflow.

Expert Matching

Humwork routes each request to a verified expert using domain fit, availability, rating, and response speed.

Full-Context Handoff

The service passes agent context to the expert and returns the answer back into the agent workflow.

Team Controls

Enterprise-facing controls include priority matching, custom expert pools, volume pricing, invoicing, and dedicated support.

Founder and Key Execs

Yash Goenka

Co-Founder and CEO (former AI engineer, two-time founder, UC Berkeley graduate, patent holder)

Rohan Datta

Co-Founder and CTO (former data scientist at MicroGrid Labs, UC Berkeley researcher, BS and MS from UC Berkeley)

Founder Force Multiplier

Goenka brings AI engineering and founder experience, while Datta brings applied data science and research training. Their mix suits a product that has to combine agent infrastructure, expert routing, and marketplace quality control.

Funding History

2025 | Founded
2026 | Backed by Y Combinator Spring 2026

Notable Open Roles

Founding Engineer

This role matters because Humwork’s core product depends on low-latency routing, MCP integrations, context handoff, and reliability inside agent workflows.

GTM Engineer

This role points to a technical selling motion where developer adoption, agent-platform partnerships, and customer-specific workflow setup may matter more than classic SaaS sales.

Competitors

Upwork:

A broad freelancer marketplace where humans hire talent, while Humwork routes agent failures to experts in real time.

Fiverr:

A task marketplace for human buyers, while Humwork is built around agent-triggered expert help.

GLG:

An expert network for human-led research calls, while Humwork targets embedded expert responses inside agent workflows.

Scale AI:

A data and human-in-the-loop AI operations provider, while Humwork focuses on live expert intervention for autonomous agents.

Invisible:

An AI operations and services platform, while Humwork positions the human layer as agent infrastructure.

Humwork

's Moat:

The likely moat is proprietary data from agent failure traces, if Humwork can turn solved human interventions into better routing and resolution quality.

How They're Leveraging AI

Expert routing for stuck agents

Humwork uses ML to match stalled AI agents with verified human experts based on the task and the expert pool. The system decides who should see the request when an agent hits ambiguity, legal nuance, code failure, or judgment-heavy work.

Agent context summarization for human handoff

Humwork uses AI to package agent context so a human expert can understand the stalled workflow fast enough to respond inside the agent loop. This matters because the expert cannot ask for a full onboarding session every time an agent gets stuck.

Expert qualification and routing feedback

Humwork uses automated assessments and interaction data to qualify experts and improve routing quality over time. The model surface is not only matching, it is measuring which humans can actually solve agent failures.

AI Use Overview:

Humwork uses ML for expert matching, question routing, content analysis, summarization, and qualification, with skill embeddings as the likely core routing layer.

Agent Infrastructure

Tasklet

Runs cloud AI agents for recurring knowledge-work workflows.

Firebase-grade infra founders are attacking workflow automation as LLMs start to operate software, with early ARR and USV as both customer and investor.

Decagon

AI agents that autonomously handle enterprise customer support across chat, email, voice, and SMS.

Agentic AI for customer support is one of the few GenAI categories with clean ROI math, and Decagon has locked in the blue-chip logo sweep that compounds into category default status.

Manus

Autonomous AI agent platform that executes multi-step business workflows end-to-end

A model-agnostic orchestration layer with real browser and shell execution captured viral enterprise demand faster than any US agent startup, a thesis validated by Meta's $2B acquisition.

Bubble Lab

Gives ops teams an AI super-employee in Slack that automates cross-platform workflows.

Zapier and Make require users to think in workflows. Bubble Lab's Pearl lives in Slack and runs multi-step operations (Jira, Stripe, Notion) from plain English, which means ops teams adopt it without learning a new tool.

Back To All Companies >