AI Agent Engineer / Applied AI Engineer (100% Hands-On Coding)
Pay: 45 LPA (20 Base) · Remote (India-based) · US Hours · 5 days/week
Note: You will own production AI systems end-to-end: agent orchestration, prompt systems, evaluations, product POCs, customer deliverables, and production feature launches. This role is designed for someone with a strong owner mindset who can work independently, perform under pressure, and deliver on tight deadlines with almost no support. If you’re a growth-minded builder, indie hacker, or cracked engineer who likes shipping fast and taking full ownership, please apply.
About UsWe’re an AI agentic learning startup based in San Francisco and Toronto , transforming how the world learns. Our platform doesn’t just stream videos—it talks back , remembers, questions, adapts, and drives 10x better retention . Picture a real-time Zoom session with your smartest teacher—except it’s fully powered by AI.
We’re backed by top-tier investors and built by a team from IIT, Microsoft, Amazon, Bain, and Goldman Sachs . If you're excited to shape zero-to-one AI products, ship fast, and build systems learners love, you’ll fit right in.
What We’re Looking For- 2–5 years of hands-on coding experience in Python and/or JavaScript/TypeScript , with strong ability to independently debug, ship, and maintain production systems.
- Deep practical experience building with LLMs, AI agents, tool use, workflows, orchestration systems, prompt pipelines, and evaluations .
- Strong understanding of agent reliability : prompt design, context management, structured outputs, tool calling, memory, retries, fallbacks, guardrails, and evals.
- Experience using or building with agent skills / tool abstractions / workflow systems extensively in production.
- Comfortable owning large model orchestration stacks , including routing across models/providers, latency/cost tradeoffs, failure handling, and observability.
- Strong product instincts—you do not just write code, you build systems that create user value.
- Proven ability to create and ship quickly —you’ve built prototypes, launched features, improved workflows, and iterated fast based on real usage.
- A track record of quantifiable outcomes —better reliability, reduced latency/cost, higher conversion, improved retention, or faster delivery.
- Nice to have: published AI/ML research , strong open-source contributions, a GitHub with meaningful traction/stars , or experience at a high-functioning startup / Series A environment .
- Available to join soon . We’re scaling quickly.
- Build and maintain AI agents in production : design, ship, and improve agent workflows that power learning experiences and internal operations.
- Own orchestration end-to-end : prompts, tools, skills, routing logic, memory, evals, monitoring, and fallback strategies.
- Design robust prompting systems : create and maintain prompt libraries, structured instructions, reusable patterns, and testing workflows.
- Run evaluations continuously : build eval datasets, define success metrics, analyze failures, and improve agent quality over time.
- Ship product features directly : work across backend, frontend, and AI layers to launch user-facing functionality fast.
- Debug everything : model failures, flaky agent behavior, broken tool usage, poor outputs, UI bugs, workflow bottlenecks, or edge-case regressions.
- Improve speed, reliability, and cost across our AI stack.
- Collaborate tightly with product, design, and learning teams to build experiences that feel intelligent, polished, and human.
- Move fast with ownership : if something is broken, unclear, or missing, you fix it.
- An engineer first —you are deeply hands-on and love coding.
- A builder/hacker who can go from idea to shipped system without waiting for perfect specs.
- Someone who is obsessed with AI quality , not just demos—reliability, evals, edge cases, and production behavior matter to you.
- A growth-minded operator who cares about shipping things that actually move user and business metrics.
- Comfortable doing whatever the moment requires: prompt tuning, writing backend logic, fixing frontend issues, reviewing outputs, improving evals, or patching workflows.
- A gritty executor who thrives in a high-speed, high-context startup environment.
- Low ego, high ownership, and highly autonomous.
You’ll likely be a strong fit if you are one or more of the following:
- An AI engineer / agent engineer who has built real LLM systems in production.
- An indie hacker or growth hacker who has shipped scrappy, high-leverage products quickly.
- Someone who has worked at a strong early-stage startup and is used to chaos, speed, and ownership.
- Someone with a research bent —for example, published work in AI/ML or deep technical curiosity beyond surface-level prompting.
- An engineer with a strong GitHub/open-source track record or visible technical credibility online.
- Customer Obsession : Everything you ship must make learners go, “Whoa.”
- Be an Owner : You’re accountable from idea to impact.
- Superpumped : High energy, high agency—we solve real problems with heart.
- Bias Toward Output : Perfect is the enemy of shipped.
- Nothing But Excellence : If it’s buggy, slow, or unreliable, it doesn’t go live.
- Competitive Pay : Base salary + ESOPs + monthly bonuses tied to outcomes.
- Work-Life Balance : Health benefits, gym perks, paid time off, and birthday leave.
- Cutting-Edge AI Work : Get hands-on with LLM orchestration, evaluations, real-time AI systems, prompt engineering, and agent workflows at production scale.
- Massive Ownership : Your fingerprints will be on the architecture, the product, and the company’s core AI systems.
- If you’re the kind of engineer who can architect it, code it, evaluate it, debug it, and scale it yourself—we’d love to talk.