Senior AI Engineer

What you’ll do

• Lead end-to-end AI workflow engagements for 2–3 clients at a time — discovery, architecture, build, ship, monitor.
• Architect production AI systems using n8n, LangChain, custom Python services, and direct LLM APIs (Claude, GPT-4o, open-source models).
• Make the build-vs-buy decisions on every engagement — when n8n suffices, when custom code is needed, when fine-tuning makes sense.
• Build observability into every workflow — logs, traces, drift detection, cost monitoring, automated alerts.
• Debug production issues across LLM behavior, integration failures, and data quality. Your code runs at 2 AM whether or not you're awake.
• Mentor 2–3 junior AI engineers — code review, architecture reviews, weekly 1:1s.
• Contribute to internal tooling — our shared n8n nodes, internal AI evaluation harness, prompt libraries.

What we’re looking for

• 5+ years of software engineering experience, with at least 2 years in AI/ML or LLM-powered systems.
• Production experience with at least one major LLM provider (Anthropic, OpenAI, Google, AWS Bedrock, or open-source via vLLM/Ollama).
• Strong Python — async, type hints, testing, package management. Comfort with at least one framework: FastAPI, Pydantic, LangChain, LlamaIndex.
• Experience with workflow tools (n8n preferred, or Make, Zapier, Temporal) — and clarity on when to use them vs. custom code.
• Vector databases and embeddings experience (pgvector, Pinecone, Qdrant, Weaviate, or similar).
• Comfort with debugging non-deterministic systems — prompt engineering, eval harnesses, A/B testing LLM outputs.
• Strong written communication — you'll write architecture documents, postmortems, and client reports.

Bonus points for

• Open-source contributions to AI tooling (LangChain, LlamaIndex, n8n nodes, etc.).
• Experience fine-tuning models on domain-specific data.
• Built and maintained an internal LLM evaluation harness or eval suite.
• Background in agencies or consulting — comfort working directly with non-technical clients.
• Public presence — blog posts, conference talks, podcasts about AI engineering.

What we offer

How to apply

Process timeline

1. Step 01

   Application review

   We respond within 5 business days.

2. Step 02

   Take-home or live exercise

   Paid · 4–6 hours · You choose.

3. Step 03

   Technical deep-dive

   90 mins with our AI lead and a senior engineer.

4. Step 04

   Founder conversation

   60 mins with Vikram (CEO) and Priya (Head of AI).