Skip to content
Engineering

AI Engineer (LLM Applications)

Take LLM features past the demo — into something a business can depend on.

Team
Engineering
Type
Full-time
Experience
Senior
Location
Remote (worldwide)
Pay
$35–$55/hr
Openings
1
Apply by
25 Sept 2026
Hiring manager
Priya Raghunathan, Engineering Lead

We build AI features that go into production and stay there: retrieval systems over private documents, classification and routing pipelines, agents doing narrow well-defined jobs. The interesting work is not the prompt. It is evaluation, failure handling, cost control and knowing when a language model is the wrong tool.

What you’ll do

  • Design and ship retrieval-augmented systems over client document sets
  • Build evaluation harnesses before shipping — golden sets, regression checks, per-category metrics
  • Instrument cost and latency, and keep both inside an agreed budget
  • Design graceful degradation: what happens when the model is wrong, slow or down
  • Advise clients honestly, including when the answer is "do not use an LLM for this"
  • Keep pace with a field that changes monthly, and separate signal from launch posts

What we’re looking for

  • 4+ years in software engineering, at least 1 shipping LLM-backed features to production
  • Strong Python, including async and a real API framework
  • Hands-on with embeddings, vector stores and retrieval quality — chunking, reranking, hybrid search
  • You have built an evaluation set and used it to reject your own work
  • Understand token economics well enough to forecast a monthly bill
  • Able to explain model limitations to a non-technical stakeholder without hedging

Nice to have

  • Fine-tuning or adapter training on open-weight models
  • Experience with agent frameworks and their failure modes
  • Prompt-injection and data-exfiltration awareness in multi-tenant systems
  • Published writing or talks on applied AI
AI Engineer (LLM Applications)

Apply for this role

We read every application personally and reply within 7 working days.

Fields marked * are required.

Include the country code, e.g. +91 98765 43210.

Years of experience
  • Less than 1 year
  • 1–2 years
  • 2–4 years
  • 4–7 years
  • 7+ years

A Google Drive, Dropbox or similar link that anyone can open.

Two to four sentences is plenty.

Confidential. We never share your details or send spam.