SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 170,000 - 200,000 / annual
RelationalAI is building AI systems that teach large language models to understand enterprise business logic, semantics, and context. The company has pioneered Superalignment technology that enables LLMs to learn from private, structured enterprise data inside the data cloud, combined with relational knowledge graphs and neuro/symbolic-relational reasoners to deliver trustworthy decision intelligence.
As a Solution Engineer, you will be embedded directly in customer environments owning outcomes end-to-end. You'll work with executives, domain experts, and data teams to identify high-impact business decisions—inventory misplacement, hidden risk concentrations, fraud patterns, capacity planning—then model their world in RelationalAI's ontology, formulate reasoning problems, write PyRel code, and deploy working systems against real customer data in their Snowflake accounts.
Every engagement produces two deliverables: the working decision system the customer sees, and the platform gaps you discover and bring back to the product team. The second deliverable is why this role exists. You'll identify modelling patterns that should be primitives, constraint formulations that belong in the platform, and workarounds that should become APIs. You argue for the strongest ones, and they become product.
You'll operate with unusual autonomy: deciding what's worth building, when workarounds are acceptable versus technical debt, and when to tell a customer their real problem differs from what they asked about. You're technical enough to debug customer environments yourself rather than filing tickets.
Key responsibilities include: owning discovery through production hardening and measurement; designing and shipping decision solutions using ontologies, rules, graph analytics, optimization, and predictive models; filling platform gaps by scoping and building solutions, then pushing general versions upstream; closing the loop with Product by bringing back reproductions and specific failure modes; running technical discovery workshops and POCs to validate solvability; documenting work as you go to create reusable reference implementations; and refusing shortcuts that compound technical debt.
You thrive in ambiguity and move with intent. You're an owner who takes full accountability for outcomes, not just your slice. Your instinct facing hard problems is to open an editor, not a deck. You have high conviction but low ego—you argue hard for what you believe, are direct about what's wrong, and change your mind quickly on evidence. You're rigorous about root-causing issues and can explain both why things broke and why fixes work. You're fast in unfamiliar territory, useful within days in new codebases or domains. You have high tolerance for enterprise friction—broken data, VDI access, security reviews, politics—and route around it to keep shipping. You're impact-driven, wanting systems you build to still be running and load-bearing years later.
This role is explicitly NOT demo-and-handoff pre-sales, staff augmentation, advisory work, or a support/maintenance role. You own outcomes, not hours or ticket queues. You deploy new things; you don't maintain legacy systems.
**Requirements:**
- 5+ years building and shipping production software, including work inside customer or partner environments
- Demonstrated end-to-end ownership: personally taken something from ambiguous problem statement to production, able to walk through the entire arc including what went wrong
- Strong SQL and deep familiarity with cloud data platforms (Snowflake, BigQuery, Databricks, Redshift)
- Strong programming ability in Python; comfort with declarative or logic-style languages is advantageous
- Comfortable reading unfamiliar source code, interpreting stack traces, and debugging systems you didn't write
- Able to hold your own with both VP-level business stakeholders and staff data engineers in the same meeting
- Comfortable operating in high-autonomy, high-velocity, low-instruction environments
**Preferred Qualifications:**
- Built analytical, decision, or reasoning applications that reached production and remained operational
- Experience with optimization, constraint solving, rule engines, graph algorithms, or ML on structured data
- Semantic modelling, data pipelines, and governance in real enterprise settings
- Track record of upstream contribution—features, tools, or abstractions you built for one customer that became standard for everyone
- Prior experience in enterprise technology, AI, or analytics platforms
