SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Togal.AI is an AI-powered cloud platform for the construction industry that automates the pre-construction takeoff process, helping estimators work 80% faster. You will own the systems that convert raw signal into pipeline—prediction models, campaign infrastructure, and AI-driven personalized outreach at scale.
You'll inherit and operate live systems carrying real revenue. Your responsibilities include: building predictive models for account scoring (estimator count, seats, expected ARR) with confidence bands; running multi-channel campaigns (email, LinkedIn, voicemail) with full ownership of routing, retries, deduplication, and deliverability; generating per-rep personalization including landing pages and AI avatar video with end-to-end attribution; grounding AI in real call intelligence to feed prep docs, follow-ups, and outbound messaging; and turning signals (job changes, promotions, news, clicks) into prioritized rep actions.
You bring 5+ years in growth engineering, RevOps, technical marketing, or sales ops with direct pipeline or revenue impact. You write production code fluently in at least one language and have applied ML experience—you've built, trained, and maintained models in production and can articulate confidence levels to non-technical stakeholders. You're comfortable with REST APIs, webhooks, OAuth, rate limits, and the debugging that follows vendor API failures. You have strong SQL skills, data intuition, and hands-on experience with modern GTM stacks (HubSpot/Salesforce, Clay, Apollo, Instantly, n8n). You have practical LLM experience including prompting, function calling, and output evaluation. You prioritize pragmatism: a working v1 this week beats a perfect spec next month, and you know which problem needs which approach.
Bonus qualifications include construction/AEC background, ML research experience, frontend or backend depth, or experience scaling GTM engineering functions. This is not greenfield work—you're operating under live conditions with noisy data and imperfect labels, making judgment calls on what to fix first while being transparent about uncertainty.
