Overview
You sell Roboflow's computer vision platform to ML engineers, data scientists, and engineering leaders. Your deals start with a technical demo, move into a trial/POC, then close through a combination of technical validation and business case. You're competing against build-it-yourself, cloud platforms (AWS/GCP/Azure), and specialized point solutions.
Role Snapshot
| Aspect | Details |
|---|---|
| Role Type | Full-cycle AE (demo to close) with SDR support |
| Sales Motion | Balanced (SDR-sourced + PLG expansion) |
| Deal Complexity | Technical consultative |
| Sales Cycle | 1-3 months (can extend to 4-6 for enterprise) |
| Deal Size | $10K-75K ACV initial, expansion to $100K+ |
| Quota (est.) | $400K-600K/year |
Company Context
Stage: Series B ($63.6M raised, GV-backed, $40M round Nov 2024)
Size: 107 employees
Growth: "Best year yet" in 2025, expanding NYC office, actively hiring across GTM
Market Position: Leader in CV developer tools with strong open source presence. Competing against hyperscaler platforms and specialized annotation/training tools. 4.7 G2 rating shows strong product satisfaction.
GTM Reality
Pipeline Sources:
- 40% SDR-sourced - Outbound to ML engineers at companies with CV use cases
- 35% PLG expansion - Free tier users and open source community converting to paid
- 15% Inbound - Product signups, demo requests from website/docs traffic
- 10% Existing customer expansion - Adding teams or use cases
SDR/AE Structure: Dedicated SDRs book your meetings. You own from first demo through close and hand to AM/CSM for expansion.
SE Support: Shared SE pool for technical deep dives, custom POCs, architecture reviews on larger deals ($50K+).
Competitive Landscape
Main Competitors:
- Cloud platforms: AWS SageMaker, Google Vertex AI, Azure ML (enterprise accounts often have commitments here)
- Specialized tools: SuperAnnotate, V7 Darwin, Encord, Labelbox (feature comparison battles)
- Build in-house (common with well-funded ML teams)
- Clarifai and other cost-focused alternatives
How They Differentiate: End-to-end platform vs stitching together point solutions, developer experience, open source credibility, deployment flexibility (cloud + edge)
Common Objections: "We can build this ourselves", "We already pay for AWS/GCP", pricing (Reddit threads complain about cost), vendor lock-in concerns, "We need to evaluate 2-3 other options first"
Win Themes: Speed to production, no infrastructure maintenance, works with their existing ML stack, active community support, proven deployment scale
What You'll Actually Do
Time Breakdown
Active Deals (45%) | Demos/Discovery (30%) | Admin/Internal (25%)
Key Activities
- Technical Demos: Live product walkthroughs showing annotation, training, and deployment. Prospects often come with specific questions about model formats, API latency, edge device support. Demos are 45-60 minutes, mostly product-driven with you narrating and answering technical questions.
- Trial/POC Management: Most deals require a hands-on evaluation. You're checking in 2-3x/week, troubleshooting integration issues, connecting them with support/SEs, and trying to move them toward a commercial decision.
- Multi-threading: Starting point is often an IC engineer, but deals over $25K need manager approval and $50K+ need director/VP sign-off. You're scheduling follow-up calls to bring in decision makers and build business case.
- Procurement Navigation: Even after technical win, you're waiting on procurement, MSAs, security reviews. Larger companies move slowly. You're chasing signatures and internal champions.
- Expansion Conversations: Once a team is paying, you're looking for other use cases or teams that could use the platform.
The Honest Reality
What's Hard
- Technical buyers take weeks to evaluate during trial. Many go silent after initial excitement. You're following up repeatedly with "How's the POC going?"
- Build vs buy debate - well-funded ML teams think they can build this themselves. You're selling time-to-value, but they often want to try building first.
- Cloud platform lock-in - enterprises have AWS/GCP commitments and you're asking them to add another vendor. Procurement friction is real.
- Pricing objections - Developer tools get compared to open source alternatives. Reddit discussions show price sensitivity ($250 for 10K images seen as expensive).
- Deal slippage - "We'll start next quarter" happens constantly. CV projects get deprioritized or timelines shift.
- PLG leads vary wildly - free tier includes students, hobbyists, researchers with no budget mixed with real prospects.
What Success Looks Like
- Close 1-2 new logos per month in $10-50K range, plus 1-2 expansion deals
- Maintain 3x pipeline coverage (AEs report needing $1.5-2M pipeline to hit $500K quota)
- 30-35% win rate on qualified opportunities
- Average 60-75 day sales cycle from demo to signature
Who You're Selling To
Primary Buyers:
- ML Engineers / Computer Vision Engineers (initial users and champions)
- Engineering Managers / ML Team Leads (approval for $15-50K deals)
- Directors/VPs of Engineering or ML (approval for $50K+ deals)
What They Care About:
- Technical validation: Does it work with their models, frameworks, deployment targets?
- ROI: How much faster can they ship CV features vs building tools themselves?
- Flexibility: Not locked into proprietary formats, can export if needed
- Support: Will they get help when things break or do they get stuck in docs?
- Cost: Total cost vs cloud platform credits they already have vs engineer time to build
Requirements
- 2-4 years closing technical/developer tool deals (selling to engineers, not business buyers)
- Ability to run technical demos and answer product questions (you don't code, but you understand ML/CV workflows)
- Experience with PLG sales motion (converting free users to paid, usage-based pricing)
- Comfortable with longer evaluation cycles and technical buyers who want to test extensively
- Track record managing 15-20 active opportunities and moving technical buyers through procurement