AI Engineer

About Vispero

Vispero builds assistive technology for people who are blind or have low vision. Our products include JAWS, ZoomText, and Fusion. We are expanding our engineering team in India to grow our assistive technology, kiosk, and digital accessibility businesses.

Role Summary

The AI Engineer builds developer tools and AI services. The first half of the role is developer tooling that fits into the places developers already work: AI coding agents, the command line, and CI/CD pipelines. The second half is the AI applications and cloud services behind our products, including model gateways, authentication, and usage metering. You will also design evaluations that tell us whether our AI features work. We want someone who takes initiative, moves quickly, and uses Claude Code or Codex as a core part of how they build software.

Key Responsibilities

• Build and maintain MCP servers, CLIs, and coding-agent plugins (Claude Code, Codex) that fit into everyday developer workflows.
• Integrate these tools with CI/CD systems such as GitHub Actions.
• Use browser automation (Playwright, Puppeteer, or the Chrome DevTools Protocol directly) for automated testing and verification.
• Build and operate backend services in Python (FastAPI) and TypeScript on AWS, including ECS, DynamoDB, and related services.
• Work on AI gateway and routing infrastructure: model selection, authentication, rate limiting, and usage metering.
• Design LLM evaluation sets, run them, and report results so product decisions rest on measured quality.
• Write tests, documentation, and clear pull requests; take part in code review.
• Work with product, QA, and subject matter experts to turn requirements into shipped features.

Required Qualifications

• Bachelor's degree in Computer Science, Engineering, or a related field, or equivalent experience.
• At least 2 years of professional software engineering experience.
• Solid grounding in computer science fundamentals and the building blocks of software systems, including data structures, algorithms, operating systems, networking, and databases, and how they fit together.
• Good understanding of how web applications work, from the browser through HTTP, APIs, and backend services to the database.
• Strong TypeScript and Node.js skills.
• Daily, hands-on use of Claude Code or Codex to plan, write, test, and review code, with examples of work you have shipped this way.
• A self-starter who takes loosely defined problems through to a shipped result and raises blockers early.
• Working knowledge of Python, preferably with FastAPI or a similar framework.
• Experience building developer tools such as CLIs, SDKs, editor extensions, or CI/CD integrations.
• Hands-on experience building applications on LLM APIs, including prompt design and tool or function calling.
• Experience deploying and running services on AWS or another major cloud provider.
• Comfort with Git, automated testing, and code review.
• Clear written and spoken English for work with teams in the US and India.

Preferred Qualifications

• Has built or contributed to an MCP server or a plugin for an AI coding agent.
• Experience designing LLM evals and tracking quality across model or prompt changes.
• Experience with model gateways, API authentication, or usage-based metering.
• Understanding of the Chrome DevTools Protocol (CDP) and browser automation experience with Playwright or Puppeteer.
• Public work we can review, such as open-source contributions or published tools.

What Success Looks Like

In the first four months, you will have shipped a working feature in our developer tooling, contributed to at least one production AI service, and built or extended an evaluation suite that the team uses to make release decisions.

The position will be between 18 LPA to 24 LPA.