About the Role

Sonic is Propel's AI-native marketing platform. It pairs a chat-based agent with each marketer's brand, content, and the data sitting inside their existing marketing stack — campaign tools, product analytics, knowledge bases, communication channels — so they can move from a campaign brief to a live, executed campaign with fewer iterations and higher-quality output.

As an AI Product Engineer at Propel, you'll work across Sonic's backend and frontend. You'll write code, reason carefully about user flows, and ship AI-powered product features end-to-end. The coding itself is largely done with AI agents (Claude and equivalents); what we hire you for is judgement, product instinct, and the ability to write specifications an AI agent can execute against without ambiguity.

You will be expected to deliver from day one. There is a written growth path with defined criteria at each step.

Key Responsibilities

Build and ship the conversational surface of Sonic — the chat interface, how the agent presents its work, how it renders generated artifacts (campaigns, copy variants, analysis), how it handles file uploads, how it surfaces deep-research outputs, and how the user shifts between different modes of working with the agent.
Build the setup experience for new clients — the onboarding flow that walks a marketer through connecting their existing marketing tools, granting Sonic access to their organization's data, and configuring their workspace.
Build the system that lets Sonic feel native to each client's brand — the configurable theming layer and design-token system that turns one shared platform into something every marketer experiences as their own.
Contribute to the system that brings external knowledge into the agent's working memory — brand guidelines, past campaigns, design references, copy archives — so that what Sonic produces reflects each client's voice and context.
Author RFCs (Request for Comments) end-to-end — written design documents that specify the problem, alternatives considered, trade-offs, the decision, and what is explicitly not being built. Sharp enough that another engineer or an AI agent can build the right thing without you in the room.
Maintain strong PR, commit, and branch hygiene at sustained cadence — squashed PRs, short-lived branches, descriptive commit messages, scoped PRs (≤ 500 lines on average), and substantive reviews on peers' PRs.
Build verification rigor on the systems you own — integration tests, evaluation suites for the agent surfaces you ship, and catching AI-generated regressions before they reach production.
Instrument latency and observability for at least one critical path you own — for example, time-to-first-token for an agent flow — with a dashboard the team can use to monitor it.
Ship through AI agents. Write specifications Claude (or equivalent) can execute against, review the output critically, and ship the result. Propel engineers do not measure their value in lines of code typed.
Collaborate closely with the designer — translate UX specifications into shipped features, and route ambiguities back rather than resolving them on your own.
Reach into core code without breaking it — schema, routing, orchestration. Past evidence of doing this responsibly in a complex codebase is required.

Required Competencies

2–4 years of total engineering experience, with at least 1 year shipping product-shaped features that integrate with LLMs, agents, or AI APIs.
Full-stack capability — comfortable in our stack: TypeScript, Node.js, PostgreSQL with Drizzle ORM, and React.
Strong product instinct. You think in user flows, not just functions. You know what “feels finished” versus “half-done” and push for the former.
AI-augmented engineering as a first-class part of your loop, not an afterthought. If working through Claude (or equivalent) feels uncomfortable to you, this role is not a fit.
Written communication strong enough to author RFCs and durable specifications that survive contact with a fresh engineer or an AI agent.
Past evidence of reaching into core code in a complex codebase without breaking it — schema, routing, orchestration.
Smart and curious — first-principles reasoning, comfortable in ambiguity, learns the next thing fast.

Bonus: experience with chat UI, artifact rendering, or streaming agent responses; has shipped an OAuth / connector integration end-to-end (e.g., Notion, Slack, Mixpanel); has worked closely with a designer on a complex product surface; startup-scale experience (Seed–Series B).

How We Measure Performance

Performance at Propel is measured on a written, six-axis rubric. Every engineer is evaluated on the same axes — what changes between levels is the bar, not the rubric itself.

Problem framing & specification — Can you describe what we should build, what we should explicitly not build, and why?
Engineering judgement & verification — Do you instrument before you ship? Do you catch regressions before users do?
System & architectural thinking — Can you reason about the system as a whole, not just the feature you're building?
Product thinking & user affinity — Do you understand why the marketer cares about what you ship?
Ownership & delivery — When you say you own something, do you mean it end-to-end?
Collaboration & communication — Can your written work survive contact with a fresh engineer, an AI agent, or a designer who joined this week?

Each axis is scored on a 1–5 scale.

Review cadence: a written six-month checkpoint (no CTC change), and a full re-score at twelve months where promotion is considered.

Performance bonus: Up to 12% of fixed CTC, paid annually after the 12-month review, conditional on named outcomes agreed at the start of the role.

Promotion to AI Product Engineer II: Considered at 12 months. Requires a rubric average of ≥ 4.0 with no individual axis below 3.

What You Get

Total target compensation between ₹12–16 LPA — fixed base plus performance bonus, anchored to our AI-native compensation framework.
A written career path: AI Product Engineer I → AI Product Engineer II → Senior AI Product Engineer, with compensation recalibrated at every promotion.
A small, intentionally lean team where AI-augmented engineering is the operating model, not a buzzword.
A written performance framework — the same rubric for everyone, a six-month checkpoint, a twelve-month re-score, no surprises.