CollectWise — AI Agent Engineer

Type: Full-time | Hybrid (3+ days/week on-site) | Flatiron/NoMad, New York, NY Compensation: $200K–$300K + 0.25%–1% equity Hiring count: 1 Visa sponsorship: Open to visa transfers (e.g. OPT, H-1B transfers) for strong candidates; no blanket new-visa sponsorship Reports to: Sean O'Brien, Founder / CEO

About CollectWise

CollectWise automates late-stage consumer debt collection with AI. The platform handles the entire recovery process for creditors — asset research, pre-legal outreach, litigation, credit reporting, and enforcement actions like garnishments and liens — via AI agents that reportedly outperform human collectors by 2x at a fraction of the cost.

Founded: 2024 | Team size: 8 | Total funding: $7M (YC + Unpopular Ventures) Industry: Financial Services, AI, Fintech Website: www.collectwise.com Office: NoMad, New York, NY

Data conflict: The Paraform company-info panel lists total funding as $500,000, but the role highlights and the intake call both state $7M. $7M used here. Confirm.

Why Candidates Should Join

• Own the whole stack: Build the entire voice AI infrastructure from scratch on LiveKit — powering ~10M calls/month in production, with full autonomy and no management layers.
• Early-stage upside: YC-backed (Fall '24), ~$2.5M ARR in under 18 months, targeting $10M ARR next year. 0.25%–1% equity.
• Real PMF in a large market: AI agents already beating human collectors 2x in a ~$35B US market.
• Direct founder access: 8-person team, reporting directly to the founder/CTO — high impact per head.

Intake Call Summary

• Automates debt collection with AI agents; ~$2.5M ARR, $7M funding, targeting $10M next year.
• Hiring an AI engineer to build in-house voice AI infrastructure with complete ownership of a system handling 10M calls/month.
• Team of 8, mostly engineers, reporting directly to founder or CTO.
• Requirements: 3–7 yrs voice AI frameworks (LiveKit preferred); AWS, Node.js/JavaScript essential; founding/startup experience a plus.
• Comp $200K–$300K incl. equity; hybrid 3 days in-office; no visa sponsorship available (see conflict note).
• Technical bar: building voice AI at scale (LiveKit or similar); prompt engineering and conversation design.
• Ideal profile: strong voice AI background, reputable companies or top universities a plus; native / highly proficient English with strong written communication.
• Interview: initial screen paid take-home technical interview with CTO.
• Pain point: low volume of qualified voice-AI candidates in prior sourcing.

The Role

Build CollectWise's entire voice AI infrastructure in-house on LiveKit, owning end-to-end development of a voice orchestration framework powering ~10M calls/month.

What You'll Be Doing

• Build the in-house voice orchestration framework on LiveKit from scratch
• Migrate all existing AI agents from the current voice provider to the new LiveKit-based system
• Own prompt engineering and conversation design for voice agents handling millions of consumer interactions
• Optimize real-time voice performance (latency, turn-taking, state management) at production scale
• Build testing and evaluation frameworks to continuously improve agent performance and business outcomes

Tech stack: Node.js, AWS, SQL, LiveKit, Voice AI, ASR/STT/TTS, LLMs, Python

Qualifications

Seniority

• 3–8 years of experience in voice AI infrastructure and backend engineering [Required]

Work Experience

• Built production voice AI systems at scale (e.g. LiveKit, similar voice frameworks) [Required]
• Built voice/agentic systems from scratch — ideally as founding engineer or at an early-stage startup [Required]
• Experience at reputable voice AI companies [Strongly preferred]
• Big Tech (Google, Meta, Roblox) OK if combined with startup or voice AI side projects [Strongly preferred]

Education

• CS degree or related technical field from a strong university [Strongly preferred]

Hard Skills

• Voice AI framework experience — LiveKit strongly preferred [Must have]
• Node.js / JavaScript and AWS proficiency [Required]
• Prompt engineering and conversational flow design for voice agents [Strongly preferred]

Soft Skills

• Strong written communication — needed for prompt engineering and conversation design work [Required]

Miscellaneous

• Based in NYC — hybrid 3+ days/week in Flatiron office [Required]

Data conflict: Seniority badge reads 3–8 years [Required]; the role description and intake call both say 3–7 years. Scoring against 3–8 pending confirmation.

Traits to Avoid

• No voice AI or agentic experience — pure backend/SWE with zero voice/conversational AI

Role Details

• Salary | $200K–$300K
• Equity | 0.25%–1%
• On-site policy | Hybrid, 3+ days/week in Flatiron/NoMad office
• Visa sponsorship | Open to visa transfers (e.g. OPT, H-1B transfers) for strong candidates (David, Aug 21 2026) — do not disqualify a strong candidate on transfer status; no blanket new-visa sponsorship
• Employment type | Full-time
• Location | New York, NY (Flatiron/NoMad)

Screening Questions

1. Can you be on-site in NYC (Flatiron, 3+ days/week)? If not currently local, are you willing to relocate?
2. What is your salary expectation?
3. How actively are you exploring new opportunities right now?

Interview Process

Stage 1 — Submit Candidate After submission, you'll be notified if the hiring manager wants to proceed.

Stage 2 — Initial Screen with Sean, Founder (30 min) First-round interview covering overall experience, background, and role/company Q&A. Conducted by Sean O'Brien, Founder.

Stage 3 — Paid Take-Home Assignment (3–5 hrs) Paid technical assignment building a project relevant to the role (voice AI / LiveKit infrastructure). Candidate is compensated for their time.

Stage 4 — Technical Interview with CTO (60 min) Technical depth plus review/discussion of the take-home. Evaluates voice AI engineering skills, system design thinking, and approach to the assignment.

Stage 5 — Offer Extended

Stage 6 — Candidate Hired

Ideal Companies & Backgrounds

Updated Jul 29, 2026

Reputable Voice AI companies Sierra, Decagon, LiveKit, Deepgram, ElevenLabs, Retell AI, Vapi

Big Tech with strong engineering talent (built voice AI projects) Google, Meta, Roblox, Apple, Amazon, Microsoft, OpenAI, Anthropic

High-growth startups / YC companies with voice/AI infrastructure focus Brex, Replit, Vercel, Resemble AI, Hume AI, Cartesia, AssemblyAI, Speechmatics, Rime Labs

Conversational AI and agent platform companies Parloa, PolyAI, Kore.ai, Voiceflow, Thoughtly, Air AI, Observe.AI, Dialpad, Five9

Non-Ideal Companies — Do Not Source Large companies where candidates spent their whole career with no voice AI / agentic experience: Google, Meta, Amazon, Microsoft, Apple, Oracle, IBM, Salesforce, Cisco, Intel

Note: Google, Meta, Amazon, Microsoft, and Apple appear on both lists — acceptable only when paired with startup or voice AI work; a full career at these with zero voice/agentic experience is the disqualifier.

Ideal Candidate Profiles

For reference only — do not source these specific profiles.

Brian Yin — LinkedIn SWE @ LiveKit, Ex-Roblox | Champaign, IL

• At LiveKit — knows the framework better than almost anyone
• Significant voice AI experience plus reputable-company background (Meta intern, strong university)
• Built voice tooling for AI agents at Roblox

Danny Lee — LinkedIn Building @ Sierra | United States

• At Sierra AI, which Sean named as a top reputable voice AI company to source from
• Direct voice AI company experience — the top priority for this role

Rejected Candidate Feedback

• Lacks sufficient relevant experience — two candidates rejected at HM Review (Aug 11, 2026)
• Unexplained experience gap — rejected at Initial Screen with Sean (Aug 11, 2026)