Senior AI Engineer (Remote)

This role is part of the Jobright TNT - the private hiring network connecting top talent with top AI startups like Perplexity, Mercor, Cresta, Suno and 150 more.

This is not a mass job posting. Only select, high-signal candidates are invited to Jobright TNT and recommended directly to hiring teams

Hiring Company: Deduction.com

One-liner: Deduction.com is building the AI tax accountant for consumers, making expert tax advice, planning, and filing available year-round without the cost or wait of a traditional CPA.

Salary: $175K/yr - $250K/yr

Why Join Us:

• Building one of the most compelling AI consumer fintech products: the first AI tax accountant for everyday consumers.

• $9M funded by top investors and angels from OpenAI, Robinhood, and Rocket Lawyer

• Founded by Sai Dhanak, a serial entrepreneur with 2 prior exits who helped scale Latch from seed to NASDAQ.

• We’re a small, high-ownership team. Engineers here don’t just implement, they define systems, make decisions, and own outcomes end-to-end.

Role Responsibilities

• Architect & Scale AI Systems: Work closely with the Founding AI Engineer to evolve our AI systems to a robust, governed, cutting-edge platform

• Harden the Reasoning & Context Layer: Design sophisticated context engineering systems—managing memory and state while improving reliability, performance, and user experience

• Build a Multi-Layered Evaluation Platform: Design and implement a robust system for continuous quality improvement, leveraging comprehensive offline test suites, LLM-as-a-judge scoring (AI Auditor), and production-based CPA moderation to create a data flywheel

• Full-Stack AI Delivery: Write production-ready TypeScript (Node.js), ensuring AI features are seamlessly integrated into our core fintech platform

• Security-First Design: Ensure context boundaries and data retrieval paths natively enforce strict security, privacy, and compliance guardrails to protect client data

• Lead AI-Native Development: Set the standard for using AI to build AI—leveraging advanced coding assistants and automated testing to maintain high velocity and socializing best practices across the team

Qualifications

Required

• Production AI Experience: You have a track record of taking AI features beyond ‘demo-ware’ and into production at scale

• Context Engineering & AI Orchestration Mastery: Proven experience building systems where the primary challenge is managing context, memory, state at scale

• Collaborative Architect: You enjoy working closely with other senior leads to brainstorm, peer-review, and iterate on complex system designs

• The Stack: TypeScript (Node.js). Production experience with AI orchestration and evaluation. Familiarity with observability tooling is a plus

• Fintech/Regulated Context: You understand that in tax, ‘mostly right’ is wrong. You have the discipline required for building in a regulated space

• Growth Mindset: You are ready to go from zero → one → scale with minimal friction, leaning into the evolving nature of the AI stack

• Early-Stage Appetite: You thrive in environments where priorities shift fast, scope isn’t always locked. You’d rather ship, experiment, and iterate than wait for a perfect spec

• You actively use AI tools as a core part of your development workflow (not optional)

• You believe the software engineer’s job is fundamentally changing—and you lean into that shift

How can I join Jobright TNT:

If this is your first time applying to a Jobright TNT role, the process works as follows:

1. Apply to your first Jobright TNT role

2. We review your background to determine if you meet the TNT quality bar

3. If qualified, your application is directly recommended to the employer

4. Once accepted into TNT, you may be:

• Invited to apply for other exclusive TNT-only roles

• Invited to private, invite-only hiring events with top startups

You will be notified of your TNT selection result.

PS: All Jobright TNT roles are 100% real, directly hired by top AI startups we partner with, and come with priority review and higher response rates than the normal application queue.