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.

Context is the moat. Everything Sonic produces is only as good as the context it pulls — each client's brand, their creative archives, the live data inside their marketing stack, and increasingly, what Sonic has learned from its own work.

As Senior Context Engineer at Propel, you will own that surface end-to-end. The pipelines that bring brand assets, design language, copy archives, images, and live data from each client's marketing stack into Sonic. The storage and retrieval that make all of it activatable in milliseconds. The evaluation framework that proves the whole system is doing its job. You build it, you ship it, you monitor it.

Your scope ends at a clean API. How that context is used inside the agent's prompts is owned by engineers downstream — you produce a high-quality, low-latency context surface; they decide how to weave it into the agent's reasoning.

We are hiring for trajectory. You join as Senior Context Engineer and grow into Staff Context Engineer scope within roughly six months — leading the architecture across all of context, with junior engineers contributing under your direction.

Key Responsibilities

Build and own the system that captures everything Sonic knows about each client's brand — brand guidelines, design language, copy archives, past creative, and other textual assets — and makes it queryable in milliseconds.
Build and own the system that pulls live data from each client's existing marketing stack — campaign tools (e.g., Braze, Customer.io), product analytics (e.g., Mixpanel, Amplitude), knowledge bases (e.g., Notion), communication channels (e.g., Slack), and behavioral data (e.g., Fathom) — and normalizes it into context the agent can reason about. OAuth, rate limits, schema drift, and new integrations as the product grows.
Build and own the system that brings non-textual context into Sonic — images today, with video and other modalities to follow — through embedding and retrieval choices appropriate to each modality.
Build the loop that lets Sonic learn from itself — capturing what the agent did, what worked, what didn't, and the patterns across tool calls and user interactions, then feeding those learnings back as context that improves future runs.
Design the end-to-end context pipeline — ingestion, chunking, vectorization, storage, retrieval, and delivery — with predictable latency, cost, and quality.
Build the evaluation framework that measures whether context is actually doing its job — retrieval-quality metrics, agent-quality regression tests, and dashboards the team uses to catch problems before users do. You ship this yourself; you do not delegate it.
Set the architectural direction for context across the platform — patterns the team builds on, decisions documented in writing, and junior engineers contributing under your direction.
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 could build the right thing without you in the room.
Mentor a junior engineer or intern to ship-readiness on a scoped piece of the context surface.
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 engineers who consume your context API, shaping what you produce to fit how the agent actually uses it.

Required Competencies

5–7 years of total engineering experience, with at least 2–3 years on production context pipelines, retrieval-augmented systems, or retrieval at scale.
Has shipped a production retrieval or context system and run the evaluations that prove it works. Both halves are required — we will ask for the specific evaluation design and what it caught.
Hands-on builder. You can sit down and ship the evaluation framework yourself this week, not delegate it.
Strong systems thinker. Opinions on vectorization choices, chunking strategy, retrieval ranking, hybrid search, and cost / latency trade-offs — and the ability to defend them.
Has built or owned at least two integration connectors end-to-end (OAuth, rate limits, normalization).
Comfortable in our stack: TypeScript, Node.js, PostgreSQL, vector storage (pgvector or equivalent). Strong adjacent stacks accepted if the ramp story is credible.
Written communication strong enough to author RFCs that another engineer or an AI agent can build from without you in the room.
AI-augmented engineering as a first-class part of your loop. If working through Claude (or equivalent) feels uncomfortable to you, this role is not a fit.
Smart and curious — first-principles reasoning, comfortable in ambiguity, learns the next thing fast.

Bonus: experience with multimodal context (image or video embeddings, multimodal retrieval); has worked on agent / tool-calling systems and understands the evaluation question for tool calls; experience with multi-tenant context isolation; has worked with Model Context Protocol (MCP) servers; has mentored a junior engineer to ship-readiness.

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 20% of fixed CTC, paid annually after the 12-month review, conditional on named outcomes agreed at the start of the role.

Promotion to Staff Context Engineer: Considered at twelve months (and possible at six months for exceptional trajectory). Requires a rubric average of ≥ 4.5 with no individual axis below 4.

What You Get

Total target compensation between ₹35–45 LPA — fixed base plus performance bonus, anchored to our AI-native compensation framework.
A written career path: Senior Context Engineer → Staff Context Engineer, with compensation recalibrated at promotion.
Meaningful equity at founding-team scale, discussed at offer stage.
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.