We’re an AI services firm that helps companies figure out where and how to use AI. We’ve built Casper Studios to roughly 45 people, worked with 30+ clients, and done deals with some of the largest companies, PE funds, and model providers.
It all comes down to how we work with clients. Before we build anything, we listen, understand the business, and help them figure out the highest-leverage way to get started on their AI journey.
We're looking for engineers in the Philippines who are AI-native, have real fundamentals in programming, data, and systems, and have the judgment to get the most out of LLMs - and know where they break. The work spans enterprise document processing, workflow automation, computer vision, and financial research tooling, across finance, healthcare, and enterprise clients.
This is a role for someone who ships, owns outcomes, and wants unusual amounts of responsibility early.
Own AI product builds end to end, from concept through production, on a client engagement
Decide what to build and how: gather requirements, pressure-test what stakeholders ask for, and prioritize the work that matters
Prototype fast with LLMs, then harden into production with the data pipelines, integrations, and evals that make it trustworthy
Do the data engineering enterprise work requires: ingestion, data modeling, and ETL
Serve as the primary technical contact for clients - talk shop with their engineers and give their executives clarity
Instrument what you build (analytics, funnel metrics, error analysis) and iterate on real usage, not assumptions
Own reliability, security, and cost: auth, secrets, PII handling, and not blowing up the cloud bill
Write up what you learn; for the team, for clients, and publicly
Strong engineering fundamentals: programming, debugging, and system design. You think well beyond the happy path.
You've shipped something real with an LLM that got actual usage - ideally with error analysis and evals on your outputs
Data engineering: data modeling and architecture, plus ETL/pipeline experience (Airflow, Dagster, Inngest, Prefect, or similar)
Web and infra fundamentals: auth and web security, profiling slow queries (N+1, unnecessary joins), CI/CD, and cost-aware deployment on a major cloud
You use modern AI tooling (e.g., Claude Code) daily, with customized workflows
High agency: you frame ambiguous problems, state your assumptions, and push work forward without being managed
Strong writing and a high say:do ratio - you can turn a long, meandering client call into a clean set of tickets
Fluency in and preference for TypeScript - much of our stack is full TS
Depth in one of our core verticals: financial services, healthcare, or enterprise / contact-center AI
Comfort being client-facing at a senior level
You've built observability for an AI product
You're plugged into the applied-AI community
You've built and deployed something 0→1; shipped it, got real usage, and iterated
You're T-shaped: deep in one area (ML, full-stack, data, or security) and rounding out the rest
You've been the solo or founding engineer on a product and owned outcomes, not just tickets
Most engineers land on one side of a line. Strong classical engineers often haven't built the judgment to know what LLMs can and can't do. Fast "vibe coders" can spin up a demo but can't debug it once it hits real data, real scale, or a real security requirement. We need both - plus the product sense and agency to run a client build largely alone. Someone who needs clean requirements and heavy supervision won't thrive here, and neither will someone whose work falls apart the moment it leaves the happy path.