收入运营与AI系统工程师
Revenue Operations and AI Systems Engineer
关于Datahash
Datahash是一个以隐私为中心的一方数据平台。我们的核心产品Datahash Signals，为营销人员提供了在网页、应用、电商、CRM、营销自动化和数据仓库与主要广告渠道（如Meta、Google、TikTok、Snap、Amazon等）之间低代码、无代码的集成方式。我们在这些渠道中是少数获得认证的集成合作伙伴之一，并直接与它们合作开展一方数据采用计划，其中广告主正在转向服务器端集成、转化API和隐私安全的测量方式。我们是一家自筹资金、盈利的公司，业务覆盖迪拜、班加罗尔、孟买和伦敦，客户遍布欧洲、中东、非洲、亚太和美国。
职位描述
你将构建并运行收入系统：在CRM记录上添加引擎标签，使收入可由引擎和门控进行衡量；仪表板；续费和催收自动化；在任何报价之前必须存在的配额模型，以及不断增长的执行常规增长和账户管理工作的代理程序。你工作的一半是收入运营；另一半是在我们的Claude和Zoho堆栈上构建代理程序，已有技能目录，并有明确的优先级计划。如果今年只招聘一人，那一定是你。
你将负责
·
对收入进行度量。在入职两周内为每笔交易添加引擎标签；确保收益看板在不同模块间保持一致；补充地理信息；编码交易估值和节奏规则，而不是依赖记忆。
- 自动化资金循环。在T-90时生成续费交易，在六周内实现催收和超额警报功能，你在第一个季度内将支付失败率降低50%。
- 先建模再承诺。在提供报价前，为每个席位建立配额与容量模型；提供每个席位的提升报告。
- 构建代理程序。负责在增长和账户管理中代理程序的构建队列：技能、评估集、批评者、决策记录、接受度遥测，在Agent SDK主机上与CTO协作。
- 保持CRM整洁。通过每日自动清理而非每季度手动处理来维护CRM卫生；发现并修复重复提案、过期日期、停用所有者等问题。
成功的表现
· 引擎标签在两周内上线，催收自动化在六周内实现，支付失败率在第一个季度下降50%。
- 本年度提供的所有配额都已提前建模。
- 代理接受度清单上线，并每周被每个职能负责人查看。
我们如何招聘
我们行动迅速，并且透明地说明我们的决策过程。一个AI助手
查看英文原文
About Datahash
Datahash is a privacy-centric first-party data platform. Our core product, Datahash Signals, gives marketers low-code, no-code integrations between web, app, e-commerce, CRM, marketing automation and data warehouses on one side and the major ad channels on the other: Meta, Google, TikTok, Snap, Amazon and more. We are one of a handful of badged integration partners across these channels and work with them directly on first-party data adoption programmes, where advertisers are moving to server-side integrations, conversions APIs and privacy-safe measurement. We are bootstrapped, profitable, and run across Dubai, Bangalore, Mumbai and London with clients in EMEA, India, Africa, APAC and the US.
The role
You build and run the revenue system: the engine tags on CRM records that make revenue measurable by engine and gate, the dashboards, the renewal and dunning automation, the quota models that must exist before any offer is made, and the growing set of agents that carry routine growth and account-management work. Half your job is revenue operations; half is building agents on our Claude and Zoho stack, with a catalogue of skills already written and a plan for which agents come first. If only one hire happened this year it would be this one.
What you will do
- Instrument revenue. Engine tags on every deal within two weeks of joining; the revenue board consistent between tiles and tables; geography backfill; the deal-valuation and pace rules encoded, not remembered.
- Automate the money loop. Renewal deals spawned at T-90, dunning and overage alerts live inside six weeks, payment failures down by half in your first quarter.
- Model before promising. Quota-versus-capacity models for every seat before an offer is made; per-seat lift reports.
- Build the agents. Own the agent build queue across growth and account management: skills, eval sets, critics, decision records, acceptance telemetry, on the Agent SDK host with the CTO.
- Keep the spine clean. CRM hygiene as a nightly automated sweep, not a quarterly cleanup; duplicate proposals, stale dates, deactivated owners, all surfaced and fixed.
What success looks like
- Engine tag live in two weeks, dunning automation in six, payment failures down 50 percent in the first quarter.
- Every quota offered in the year was modelled first.
- Agent acceptance ledger live and read weekly by every function owner.
How we hire
We move fast and we are transparent about how we decide. An AI assistant helps us screen applications against the must-haves in this description and helps schedule conversations; every decision about a person is made by a person.
Step 1: a short structured screen within two working days of applying. Step 2: a 50-minute conversation with the hiring manager. Step 3: a small case exercise close to the real work, which you present to two or three of us. Step 4: fit and terms with our VP Delivery or Head of Sales and Partnerships, and for senior roles the founder. We aim to give you a decision within five working days of the final conversation.
We evaluate evidence rather than claims, the quality of your questions, how you handle ambiguity, and how naturally you already use AI in your own work.
Requirements
What we need from you
- Four to five years in revenue operations, sales operations, business systems or analytics engineering in a B2B software company.
- Hands-on with a CRM as a builder (Zoho preferred; Salesforce or HubSpot fine), with workflows, functions, webhooks and reporting.
- You write code: Python or JavaScript, SQL, APIs; you have shipped automations that other people relied on.
- You have built with LLM tooling (Claude, agent SDKs, MCP or similar) beyond chat, and can explain what you would never let an agent do.
- Rigour with data: you distinguish gross from net, subscription from invoice, and you check both directions before you trust a number.
Nice to have
- Zoho One depth: CRM, Books, Billing, Desk, Flow, Analytics.
- Experience in a partner-led or usage-billed business.
- Experience designing evals or QA for AI systems.
Probably not a fit if
- You want to be an analyst producing reports; this seat builds systems that produce them.
- You are uncomfortable owning production automation with a kill switch.
Benefits
- A startup environment backed by a stable, bootstrapped, profitable business.
- Competitive compensation and one of the most employee-friendly ESOP programmes in the industry.
- Work from anywhere, on your hours; coworking access when you want a change from home.
- Health cover for you and your family.
- Real international exposure: clients and partners across the Gulf, Europe, Africa, India, APAC and the US.
- A yearly offsite that we take seriously, and a culture that does not.
Datahash is an equal-opportunity employer. We hire on evidence and fit for the work and do not ask for salary history. Compensation is discussed at the fit-and-terms stage.
Originally posted on Himalayas