Forward Deployed Engineer
Forward Deployed Engineer
About us:
We build the governed context layer for the agentic enterprise. We take complex, heterogeneous enterprise data and auto-generate knowledge graphs from it — no manual schema design, no ontology engineering. Those graphs power LLM agents that automate workflows that previously required skilled human judgement and hours of manual work.
The role
We sell outcomes, not technology. The core engineering team builds the product. You deploy it. You work at client sites on real problems: validating auto-generated graphs, identifying what workflows are worth automating, building the agent prototype that makes it tangible, and enabling implementation partners to take it into production.
It's a technical role with a client-facing orientation. Equally comfortable reading a graph query and presenting to a CTO. Not a demo engineer — someone who can configure the product, break it, fix it, and explain why it matters to someone who doesn't know what a triple is.
Verticals
Pharma & Health: Competitive intelligence synthesis, KOL engagement routing, MLR risk flagging, regulatory change impact assessment. Clinical data, market signals, HCP networks, and regulatory history connected in one graph.
Finance: Product-to-client matching against versioned eligibility rules, multi-hop contract reasoning, governed advisory with full audit trail. Products, rules, and decisions connected across silos.
What you'll own
1. Getting the graph live. Figure out which connectors a client needs, work with the core team to configure them, inspect the auto-generated graph for correctness — entity coverage, relationship accuracy, gaps that'd break a downstream agent.
2. Mapping what's worth automating. Find the workflows that are slow or error-prone because the right context isn't accessible at decision time. Graph analytics surfaces the structure; the goal is always the workflow it unlocks.
3. Building the proof of value. A working prototype on real data — credible enough to get a yes, specific enough to hand off. Integrated with the client's existing LLM. Runs live. Tells a clear story.
4. Enabling partners. Make the technical handoff to implementation partners work. They build the production system. You make sure they have everything they need to do it without us in the room.
What you'll bring
· Domain knowledge. Real experience in pharma, finance, or both. Knows what's worth automating and why.
· Graph literacy. Can query and inspect graph structure. Knows whether a generated graph is fit for a workflow.
· Agent fluency. Has built agents that do something real. Knows what goes wrong and why.
· Integration thinking. Can map a client's data environment to the connectors needed to bring it into the graph.
· Demo craft. Builds things that work live. Reliable, clear, polished enough for a sales process.
· Client communication. Holds a room of senior stakeholders. Moves between technical and commercial without losing either audience.
Nice to have
· Workflow automation in pharma or finance
· GraphRAG and multi-hop reasoning
· Graph quality assessment
· SI / implementation partner experience
· Regulatory data fluency (MiFID II, MLR, ACPR)
· Prior FDE or technical consulting background
· Streamlit or equivalent rapid prototyping
Why this role
Most enterprise AI deployments stall between the model and the workflow. The data's messy, the context is missing, and no one's figured out how to make the agent reliably do the thing the business actually needs. That's exactly the problem this role exists to solve — and you'll be solving it across some of the most complex data environments in pharma and finance, with a product that's built specifically for it. Architectural input, real client problems, and outcomes you can point to.