Free · takes a minute · your resume fills your profile
CoreHRx never asks candidates for money. If anyone does, .
Sign in to see your fit: the skills, experience and location this job asks for, checked against your profile.
Location: Remote US
Department: Technology
ABOUT OUTMARKET
Outmarket is the AI platform for insurance, trusted by more than 250 brokerages to run the work their business depends on. Commercial insurance still runs on dense documents and slow, manual workflows, and that is exactly what we automate: quote comparisons, coverage gap and tower analysis, policy review, and proposal generation, all grounded in our customers’ own data and source-cited so teams can trust the output.
The impact is concrete. Teams save 12 to 15 hours per person every week, cut errors by roughly 65 percent, and win more business, all on infrastructure that is SOC 2 Type II certified, single-tenant, and never used to train AI models. We are an AI-first company in both what we build and how we work, shipping quickly and in close partnership with the agencies that rely on us.
WHAT YOU’LL GET
A high-impact role with ownership from day one.
Competitive compensation and meaningful equity.
Direct collaboration with founders and real users.
Remote-first flexibility.
The opportunity to help build an AI-native product from the ground up.
ABOUT THE ROLE
We are hiring a Software Engineer to build AI-powered features across our product and technical stack. You will work on real customer workflows, from AI and backend systems to product experiences, with strong support from a senior team.
WHY THIS ROLE
Build features that directly affect real customer workflows and outcomes.
Work across a broad technical stack in an AI-native environment.
Grow quickly with mentorship from founders and experienced engineers.
WHAT YOU’LL DO
Build product features powered by LLMs, including retrieval-augmented generation (RAG), embeddings and semantic search, prompt design, structured output, and tool-using agents.
Work hands-on with the mechanics of AI systems: chunking and embedding documents, vector stores and retrieval, context construction, and evaluating output quality.
Build the backend services, data pipelines, and product UI that turn these AI capabilities into reliable, production-grade features.
Turn messy real-world insurance data into trustworthy, source-cited product capabilities.
Collaborate across product, sales, and customer success to ship high-leverage work.
WHAT WE’RE LOOKING FOR
1 to 3 years of software engineering experience (or equivalent), with strong fundamentals.
Proficiency in Python and/or TypeScript and comfort with SQL.
A real understanding of how modern AI systems work under the hood: LLMs and prompting, tokens and context windows, embeddings and vector search, RAG, and basic evaluation of model output.
Genuine interest in applied AI and a desire to go deep on the domain.
Strong ownership and comfort in a fast-moving environment.
BONUS IF YOU HAVE
Hands-on experience with LLM orchestration and RAG tooling (for example LangChain, LlamaIndex, or similar) and vector databases (for example pgvector, Pinecone, or Weaviate).
Experience building agents, tool/function calling, or evaluation and prompt-testing pipelines.
Experience with React, Postgres, or cloud platforms, or with document-heavy data.
From live job posts on CoreHRx, updated 10 Nov 2026. Skill figures count posts that list their skills; salary figures only posts that show a salary.
The post does not state the pay. As an estimate, comparable posts on CoreHRx advertise ₹11L – ₹24L a year (typical for IT Jobs in India, 818 posts).
The post asks for 1–3 Yrs experience.
It is a remote job.
It is a full time job.
Interview preparation · 60 questions
This guide is for Indian job seekers preparing for SQL rounds in data analyst, data engineer, backend developer and database roles, from freshers to experienced professionals. It has 60 questions in three levels: easy covers joins, grouping and NULLs; medium covers window functions, CTEs, indexes and isolation levels; hard covers locking, MVCC, query tuning and classic puzzles such as login streaks and medians. Answers use ANSI SQL, with MySQL 8 and PostgreSQL notes where the two differ. Write each query against your own sample tables first, then compare your result with the sample answer.
Questions you may be asked
Create a free profile and we email you new IT jobs you can do remotely, at most once a week.
