Founding Engineer (AI + Backend) - Remote, India

present of AI; India; Remote; INR 900,000 - 1,200,000; Full time

Job summary

present of AI is seeking a backend-oriented engineer to design benchmarks from scratch and build scalable evaluation systems that run largely unsupervised. You will decide what to measure, how to measure it, and ensure results stay trustworthy as products evolve.

The role sits near agent tool use, automating API calls, scoring, drift checks, while you frame the questions and ground them in real-world tasks. Ownership of big pieces from day one is expected.

Job description

The role asks for someone to design benchmarks from scratch and build the systems that run them at scale, largely unsupervised. A strong candidate brings solid backend engineering, hands on work with large language models as judges or evaluators, and the judgment to decide what is worth measuring and how to keep results trustworthy as products change. Ownership of big pieces from day one is expected.

How it fits the AI era

The work sits close to how agents pick tools, so much of it is building the evaluation layer that agents lean on. Automation handles running vendor APIs, scoring and drift checks, while the human part is framing the right questions, grounding them and judging what holds up. Model behavior research stays a person led activity even as the pipelines run themselves.

How it builds AI experience

A person here would gain direct practice in eval design, LLM as judge pipelines, synthetic data generation and drift detection, all aimed at how agents compare and choose tools. Building proof of small reproducible benchmarks or scoring systems for developer tools would show the right instincts for this company.

Worth showing

• Benchmark design from scratch
• LLM as judge pipelines
• Large scale backend systems

This is our own read, written from the employer's posting and the company's Spotlight. The employer's posting is the source of truth for the role.

Openbenchmarks runs independent, open-source benchmarks that compare APIs, tools and agents on the same tasks under the same conditions, per the company.