Forward Deployed Engineer

Overview

You're the technical bridge between Glimpse's product and customer realities. You work with CPG brands to integrate Glimpse into their ERP systems (SAP, Oracle, NetSuite), connect to retailer data portals, and customize the AI to handle their specific deduction types. Part of your time is pre-sales (scoping technical requirements, running proofs of concept) and part is post-sales (implementation, troubleshooting, optimization). You're comfortable writing code, analyzing data, and explaining complex technical concepts to non-technical finance teams.

Role Snapshot

Company Context

Stage: Series A (estimated)

Size: ~45 employees, scaling FDE function as they move upmarket

Growth: Signed largest customer ever in January - likely a complex enterprise deal that needs dedicated FDE support

Market Position: Building out FDE function now as they scale with larger, more complex customers

GTM Reality

Pipeline Sources:

• You're pulled into deals by AEs when technical scoping is needed (usually mid-late stage)
• Proactive outreach to customers for optimization projects

SDR/AE Structure: AEs bring you in during sales cycles, you work with CSMs post-sale

SE Support: You ARE the SE support - building out this function

Competitive Landscape

Main Competitors: Custom in-house solutions, legacy deductions software, manual processes

How They Differentiate: AI that actually works with their specific data formats and retailer portals - your job is proving that

Common Objections: "Our data is too messy", "Our ERP is too old", "Our retailers use weird formats", "Implementation will take forever"

Win Themes: You can show it working with their actual data in a POC, faster time-to-value than building in-house, handles edge cases that manual processes miss

What You'll Actually Do

Time Breakdown

Implementations (40%) | Pre-sales scoping/POCs (25%) | Troubleshooting (20%) | Internal product feedback (15%)

Key Activities

• Pre-sales technical discovery: Joining AE calls to understand the customer's tech stack, data formats, ERP system, and retailer integrations. You're asking questions about API access, data volumes, latency requirements, and security constraints. Then scoping level of effort for implementation.
• Proof of concepts: Taking a sample of their real deductions data, running it through Glimpse, and showing it works with their specific retailer formats and deduction types. This usually involves custom scripting to ingest their data, tuning the AI models, and presenting results that prove accuracy.
• Implementation work: After the deal closes, you're building the actual integrations - connecting to their ERP, setting up data pipelines from retailer portals, configuring the AI to handle their specific deduction codes and business rules. This might involve writing Python scripts, SQL queries, API integrations.
• On-site time: Spending days or weeks embedded with customers during critical implementation phases. You're sitting with their finance team, understanding their workflow, debugging data issues, and training them on the platform.
• Post-launch optimization: Once live, you're monitoring performance, fixing misclassifications, tuning AI models based on their feedback, and iterating to improve accuracy and coverage.

The Honest Reality

What's Hard

• Customer data is always messier than expected - missing fields, inconsistent formats, undocumented business logic. You spend a lot of time cleaning and normalizing data before you can even start the real work.
• Legacy ERP systems are painful - APIs are poorly documented, data extraction is slow, and customers' IT teams are stretched thin. You're often blocked waiting for access or approvals.
• Retailers don't standardize anything - each retailer (Walmart, Target, Kroger) has their own portal, data format, and deduction codes. You're building custom logic for each one.
• Scope creep is real - customers ask for "just one more thing" during implementation, and saying no can hurt the relationship. You're constantly negotiating scope vs timeline.
• You're on-call for critical issues - if the system breaks or misclassifies a $100K deduction, you need to jump in fast. Finance teams don't tolerate downtime.
• Travel can be heavy depending on customer location and deal complexity - expect 20-40% travel to customer sites.

What Success Looks Like

• Implementations completed on time or early (rare but possible with good scoping)
• 90%+ AI accuracy on deduction classification within 60 days of launch
• Customers seeing ROI within first quarter post-launch
• Low escalation volume - customers trust the system and don't need constant hand-holding
• Product feedback incorporated into roadmap - you're influencing what gets built based on real customer needs

Who You're Selling To

Primary Buyers:

• Finance and operations teams (day-to-day users)
• IT teams (integration approvals and access)
• CFOs and finance directors (care about risk and reliability)

What They Care About:

• Will this actually work with our messy data? (Proof via POC)
• How long until we're live? (Implementation timeline)
• What happens when it breaks? (Support and SLAs)
• Can we trust the AI's decisions? (Accuracy and auditability)
• How much will this disrupt our current process? (Change management)

Requirements

• 3-5 years engineering experience, preferably in data engineering, integrations, or customer-facing technical roles
• Strong coding skills in Python (primary), SQL, and API integrations
• Experience with ERPs (SAP, Oracle, NetSuite) or financial systems
• Comfortable with ambiguous, messy data problems - you're not building greenfield systems, you're making existing chaos work
• Can communicate technical concepts to non-technical audiences
• Willing to travel 20-40% for on-site customer work
• Startup mentality - comfortable figuring things out without a playbook