What you'll own
Production-grade Python ingestion pipelines for a global data marketplace operating at scale: 120M+ files, 1.4 TB of data delivered weekly, dozens of active data products. You design and build the systems that move data reliably from ugly real-world sources into a governed, high-availability estate. Your code runs in production, not in a backlog.
How we work
Critical Propulsion consultants bring a different standard to wherever they work. Our model is built around swarm delivery and AI-amplified output with talent dense teams, and that's the lens you carry into every engagement. Your first engagement embeds you directly within a client team, where you'll adapt to their environment while holding your own bar for how fast and how well work gets done.
What we expect
- You write Python the way it was meant to be written. Class hierarchies, abstract base classes, dependency injection, polymorphism. You architect master controller / services / helpers structures and you can explain exactly why you made each design choice.
- You build RESTful API integrations that don't break. Authentication flows, pagination, error handling, retry logic. You've been burned by flaky third-party APIs before and your code reflects it.
- You write SQL for real work. Querying, transformation, pipeline validation. Not just SELECT *.
- You build ingestion pipelines that hold up in production. Fault-tolerant, resilient, modular. You think about observability and traceability before you write the first line.
- You work on Azure data infrastructure without hand-holding. ADLS, Azure SQL, and the patterns that connect them.
- You operate autonomously. No status meetings that could be a message. No decks that could be a decision.
- You communicate directly. When something is broken or the spec is wrong, you say so with evidence.
- You embrace agentic development. You don't treat AI as autocomplete. You delegate real work to AI agents, review their output critically, and iterate fast.
What we don't filter on
- Years of experience as a number. If you can ship production Python pipelines in 5-day pulse cycles, the number on your resume is irrelevant.
- Specific tooling as a prerequisite. This first engagement is Python-heavy by design, but we hire to a team, not a role. Broad engineering judgment and the instinct to pick up whatever the work demands is what sets consultants apart here. Deep expertise in one thing is a starting point, not a finish line.
- Pedigree. No school or company name substitutes for demonstrated ability to build pipelines that don't fall over in production.
Nice to have
- Azure Databricks and PySpark for high-volume distributed processing scenarios.
- Familiarity with XBRL and iXBRL financial data formats.
- Git/GitHub fluency and experience working in trunk-based or short-lived branch workflows.
- Exposure to near real-time processing patterns.
What you get
- A team where everyone builds. No layers of management between you and the work.
- AI agents as real teammates. You'll use AI tooling to accelerate development, testing, and documentation.
- Pulse cycles that create natural rhythm without traditional sprint overhead.
- Direct client impact. Your pipelines move real data to real consumers in days, not quarters.
- Competitive comp sized for senior consultants, not blended-team billing rates.
How to apply
Show us something you built. A pipeline design, a hard ingestion problem you solved, a code pattern you're proud of. Skip the cover letter unless you actually want to write one.
Real conversation. No fluff.
