Applied AI Engineer, Full-Stack

Tally; San Francisco (CA); On-site; USD 180,000 - 260,000; Full time

Benefits offered by this job

Competitive salary and equity

Job summary

Tally is seeking an Applied AI Engineer to own the core architecture of its billing and data platform. You’ll tackle entity resolution across multi-party data, semantic context, and multi-hop reasoning to automate complex workflows from day one.

You’ll build a context graph ingesting data from emails, PDFs, ERPs, and payment portals, and design multi-step agent pipelines with tool calling and MCP integrations to automate reconciliation and reporting.

Qualifications

• 4 years of experience in high-quality engineering environments.
• Experience shipping agent applications with tool calling, MCP, context engineering, and multi-step reasoning.
• Strong TypeScript skills and full-stack comfort.
• Experience across the AI lifecycle: data analysis, evaluation, A/B testing, and iterative improvement.
• Ability to think like a PM as well as an engineer, making smart tradeoffs.

Responsibilities

• Build a context graph that ingests data from emails, documents, ERPs, industry systems, and payment portals.
• Develop multi-step agent pipelines with tool calling and MCP integrations to automate reconciliation and reporting.
• Create evaluation harnesses that measure agent quality and drive continuous improvement across the model lifecycle.
• Develop internal APIs and interfaces powering reconciliation, reporting, and automation.
• Own production AI infrastructure with observability, reliability, and cost efficiency.

Skills

TypeScript

Full stack

Agent applications

Context engineering

A/B testing

PM mindset

Tools

Neo4j

Job description

About Tally

Over $10T is spent on global supply chain and logistics services each year, yet nearly every invoice is still created manually. The complexity has historically been too great for traditional software to automate, with dozens of parties, contracts, and exceptions on every shipment.

Tally is the first billing platform that reasons, remembers, and understands the business complexity of every transaction, helping enterprise supply chain organisations accelerate their order to cash.

At its core is our logistics context graph, a live intelligence layer that maps every party, contract, and shipment event, and reasons across them to automate order-to-cash with accuracy. Tally understands the operational context behind every transaction to ensure invoices are complete, correct, and defensible.

Our team brings deep experience across logistics operations, enterprise systems, and AI infrastructure. We’re backed by the best vertical AI investors and we’re building a small, high-agency team of engineers who want to own foundational systems in a category-defining company.

Why This Role Matters

We need an Applied AI Engineer who wants to own the hardest layer: the core architecture. The hard problems are entity resolution across inconsistent multi-party data, semantic context across inconsistent industry language, and multi-hop AI reasoning over all of it. This is foundational work with direct product impact from day one.

What You’ll Build

• A context graph that ingests and standardises data from natural language (emails), documents (PDFs, XLS), ERPs, industry operating systems, and payment portals


• Multi-step agent pipelines with tool calling and MCP integrations that automate financial reconciliation, invoice processing, and reporting


• Evaluation harnesses that measure agent quality and drive continuous improvement across the model lifecycle


• Internal APIs and interfaces powering reconciliation, reporting, and automation


• Production AI infrastructure optimised for observability, reliability, and cost efficiency

What We’re Looking For

• 4 years of experience in high-quality engineering environments; startup experience a strong plus


• Demonstrated experience building and shipping agent applications with tool calling, MCP, context engineering, and multi-step reasoning


• Strong TypeScript skills; comfort working across the full stack


• Experience owning the full AI lifecycle: data analysis, evaluation, A/B testing, and iterative improvement


• Ability to think like a PM as well as an engineer; you make smart tradeoffs and move fast


• Ability to reason about cost, latency, and reliability tradeoffs when selecting and combining models


• High ownership mindset and a bias for action

Bonus Points

• Experience with Neo4j or other graph databases


• Experience with LLM post-training, fine-tuning, or open-source model evaluation


• Interest in helping recruit and mentor future engineers as the team grows

What to Expect

You’ll work directly with the founders with full ownership over consequential systems from day one. Competitive salary and equity.