MathCo
Website:
mathco.com
Company:
https://www.linkedin.com/company/themathcompany
Seniority: Mid-Senior level
Industries: Data Infrastructure and Analytics and IT Services and IT Consulting
Job details:
About the Company
TheMathCompany or MathCo® is a global Enterprise AI and Analytics company trusted by leading Fortune 500 and Global 2000 enterprises for data-driven decision making. Founded in 2016, MathCo builds custom AI and advanced analytics solutions to solve enterprise challenges through its hybrid model. NucliOS, MathCo’s proprietary platform, enables connected intelligence at a lower total cost of ownership (TCO).
At MathCo, we foster an open, transparent, and collaborative culture, making it a great place to work. We provide exciting growth opportunities and value capabilities and attitude over experience, enabling our Mathemagicians to 'Leave a Mark'.
We are looking for a seasoned forward-deployed product and AI engineer to anchor the offshore
seat of a small, integrated global delivery team. You will own the application and intelligence layer
of AI-enabled decision solutions — decision applications, agents, evaluation harnesses, and
platform-native application surfaces — built directly on the client's enterprise platform, and you will use AI coding agents to deliver at a pace and quality bar conventional teams do not reach. The
role combines hands-on platform engineering, system design for LLM-inclusive architectures, and
the techno-functional fluency to carry business stakeholders through technical trade-offs.
Responsibilities:
System & Solution Design
- Design end-to-end solution architectures for AI-enabled decision applications and data foundations — ingestion, curated and gold layers, semantic models, and the application surface — on the client's enterprise platform.
- Design complex software systems in which LLMs are one component among many: retrieval and context pipelines, agent orchestration, evaluation harnesses, guardrails, and integration with enterprise systems — making deliberate architecture choices for accuracy, latency, cost, and failure modes, not assembling from a single vendor toolkit.
- Record significant design decisions as Architecture Decision Records (ADRs), capturing the rationale, the trade-offs considered, and the production consequence of each choice.
Techno-Functional Translation
- Act as the technical counterpart to business stakeholders: explain technology trade-offs — platform-native vs. custom, accuracy vs. cost and latency, scope vs. timeline — in business terms, and guide clients to informed decisions.
- Translate business requirements into a structured feature backlog with acceptance criteria and measurable evaluation thresholds; surface ambiguity and conflicting stakeholder aims early rather than absorbing them into scope.
- Apply working domain knowledge (CPG, Retail, Pharma, or Manufacturing) to data modeling, KPI definitions, and edge-case identification, in partnership with the onshore domain lead.
Platform-First Build
- Default to platform-native capabilities — on Databricks, Snowflake, or the relevant hyperscaler stack — for the front end, back end, and operational surface, reserving custom build for where it demonstrably earns its keep.
- Capture context gathered during the engagement in platform-native constructs (semantic layers, metric views, governed data products) so it compounds across markets and adjacent use cases rather than being rebuilt.
- Own data and access readiness at engagement start: profiling, data-quality gates, and the pipeline foundations the build depends on.
AI-Accelerated Delivery to a Production Bar
- Multiply personal throughput with AI coding agents: frame and direct multiple parallel build tracks, then specify, review, and harden what the agents produce — the quality bar is production engineering, not prototype output.
- Build production-ready from day one at the scope the engagement allows: CI/CD, versioned prompts and evals, secrets hygiene, and observability from the first sprint.
- Stand up the evaluation harness early and publish scores on a weekly cadence; demonstrate working software to the client from a correctly configured environment each week.
Delivery Discipline & Reuse
- Track delivery against committed timelines; make scope additions visible as priced scope trades rather than silent absorption.
- Run the engagement's quality gates and checklists through to sign-off; complete handover documentation — runbooks, ADR logs, operational artifacts — executable by client teams independently.
- Return reusable assets, learnings, and failure-register entries to the central library at closure, so each engagement hardens the next.
Required qualifications & experience
- 8+ years' experience in software and AI engineering, spanning application development and
- analytics, with 10+ years of overall consulting experience.
- Completed an AI Engineering Professional certification and required classes — e.g., Databricks
- Generative AI Engineer, Google Cloud Professional Machine Learning Engineer, or Azure AI Engineer.
- Professional-level certification on at least one MathCo-preferred platform — Google Cloud (including Gemini), Databricks, AWS, Azure, or Snowflake; professional tier preferred over foundational / associate.
- Minimum 6–8+ AI application projects delivered with hands-on development experience on Databricks or other MathCo-preferred platforms.
- Working knowledge of two or more common cloud ecosystems (AWS, Azure, GCP), with deep expertise in at least one.
- Deep experience building and shipping LLM-based applications — retrieval pipelines, agent orchestration, and evaluation harnesses — with working knowledge of Spark and distributed data processing.
- Familiarity with CI/CD for production deployments.
- Working knowledge of LLMOps and MLOps.
- Current knowledge across the breadth of Databricks product and platform features, particularly its generative AI and agent tooling.
- Familiarity with optimizations for performance and scalability.
- Demonstrated experience designing complex software systems, including current experience designing systems that incorporate LLMs — as distinct from building solely with managed toolkits such as Vertex AI or Azure AI Foundry.
- Domain exposure in one or more of CPG, Retail, Pharma, or Manufacturing preferred.
Click on Apply to know more.