All open roles

AI Product Manager

AI Product Manager

Intermediate · 3+ years · Remote (Algeria) · Full time

This role requires having shipped at least one AI system into production and having measured its quality after launch, regardless of the candidate's previous title. Candidates typically come from an engineering, product management, product ownership, or data background, usually with 3 to 6 years of relevant experience (3 years minimum). The strongest signal sought is prior experience proving that a probabilistic system was good enough, to a client, a regulator, or an executive, backed by data.

Responsibilities

  • Hold a position under pressure: when a client pushes back on a defined threshold, respond with an argument, not an apology.
  • Take ownership in ambiguity, turning a vague need into something buildable and testable without a handed-down spec.
  • Be measured on the value created rather than on velocity or tickets closed, specifically on whether the system meets the committed bar and changes something for its users.
  • Build and use internal tooling, including a platform for managing production prompts, an operational system carrying project state across the firm, and reusable AI skills used daily by the team. When a pattern works, it becomes something everyone can reuse.

Requirements

  • At least one AI system shipped into production, with its quality measured after launch.
  • Background in engineering, product management, product ownership, or data, typically 3 to 6 years of experience (3 years minimum).
  • Proven experience demonstrating that a probabilistic system was good enough to a client, a regulator, or an executive, supported by data.
  • Understanding of how LLM systems work: retrieval, structured outputs, agentic patterns, the quality, latency, and cost tradeoff, and how such systems fail.
  • Ability to distinguish a demo that works from a system that works.
  • Technical autonomy: ability to read and write enough Python and SQL to automate a task, run a test, and process evaluation data, with an AI coding harness handling the heavy lifting. Production code writing is not required.
  • An engineering degree or an equivalent foundation is favored, since independent iteration (for example, testing a prompt variation) without relying on an engineer is central to the role.
  • English required.

Nice to have

  • French language skills, since some client work happens in French. Internal communication, documentation, and meetings are always conducted in English.

Additional details

  • The requirements above describe the bar for a senior-level hire. A strong intermediate candidate with real potential is also welcome; an internal training program exists and is seriously invested in.
  • Employment type: full-time.
  • Location: remote (Algeria).

What happens after you apply

The same steps for every role we hire for.

  1. You apply. One form, one PDF. Your resume is read by AI to pre-fill your profile, and a person reviews all of it.
  2. A screening conversation. If your profile fits the role, we invite you to an AI-assisted screening conversation. It asks the questions and summarizes your answers with exact quotes. It does not score you and it does not decide anything.
  3. A person reads it. A real person at Betterfit reads your screening within 2 business days and writes back, whether the answer is yes or no. Every decision here is made by a human, in writing.
  4. The company sees your file. If we put you forward, the company hiring for the role sees your profile, a short note from us on why you fit, and your whole screening conversation: every question we asked and every answer you gave, in your own words. They never see what we wrote about you internally, and they never see the recording or transcript of an interview call. Your name and photo are shown only if you agreed to that.
  5. The interviews. A video call, no puzzles and no test, just a conversation about work you have actually shipped. It starts with Betterfit; if the company wants to meet you, someone from that company is on the call too. Every interview call is recorded and transcribed. We tell you the company, and who will be there, before you book anything, and you can say no.

Every placement starts with a two-week working trial. What we keep, and for how long, is in our privacy policy.

Apply for this role

One form, one PDF. A person at Betterfit reads it and writes back.

Applying for: AI Product Manager

Your resume is parsed by AI to pre-fill your profile, and a person reviews everything. If we invite you to an AI-assisted screening conversation, a person reads it and decides. No AI decides anything about your application. Once a person at Betterfit has reviewed and validated your application, the company hiring for the role sees your profile, your name, and your whole screening conversation, every question and every answer, in your own words. If another company we are hiring for has a role you suit, we may put you forward for that one as well, under the same rules. If a company invites you to an interview, it happens on a video call that is recorded and transcribed, with someone from that company present. By submitting, you agree to this and to Betterfit using your information to evaluate your application. See our privacy policy.