About Bolna
Bolna is Voice AI infrastructure built for India - and now for the world. We help businesses deploy intelligent voice agents that can call, converse, and convert in any language, at scale. From collections to customer support to sales, our agents handle millions of conversations so humans don’t have to.We’re a YC F25 company, backed by General Catalyst, with 1,050+ paying customers and growing fast. Our team of ~40 is based in Bengaluru.
About the role
This isn't a relationship-management role that escalates technical problems to someone else. Think of it as a technical product manager who also happens to own the relationship.Bolna's agents run on a stack of ASR, LLM, TTS, and telephony providers stitched together in real time - and every enterprise client is running that stack against a business outcome that actually matters to them: a payment collected, a delivery confirmed, a candidate screened.
When something breaks - a latency spike, a misfired transfer, a prompt drifting off-script under a live client's traffic - you're the first person who has to understand why. Not eventually. In enough technical depth to either fix it yourself or tell engineering exactly what's wrong.
Day to day, that looks like:
- Reading transcripts and logs, forming a hypothesis - was it ASR, the LLM, TTS, or the telephony leg - and arguing that hypothesis with the person who can fix it.
- Writing and rewriting prompts under real client constraints.
- Working closely with Bolna's own PMs and engineers - bringing them problems worth solving, backed by evidence, not a client complaint forwarded verbatim.
- Noticing when you've fixed the same thing three times manually, flagging it, and getting the product team to build it instead.
This role rewards high agency and real ownership. You don't wait to be told what to fix - you find it and drive it to resolution.If you'd rather diagnose a call failure than write a status update about one, this is the CS job you actually want.
What you'll own
- Client relationships, end-to-end. On the client side, that means working directly with their engineers, PMs, and engineering managers, not just a single business point of contact.
- Analytics & Reporting: Strong ability to analyze and visualize data, work with datasets, and derive actionable insights. Proficient in SQL for data querying and Excel for analysis, reporting, and dashboard creation.
- Technical issue triage. Latency spikes, ASR misfires, voicemail detection bugs, call-transfer failures - you diagnose root cause first, working directly with engineering, rather than passing along a client complaint verbatim.
- Translating business asks into agent behavior. When a client says "the bot should handle this differently," you own turning that into a working prompt - across whichever ASR/LLM/TTS combination that account runs on - and iterating on it as production data comes in.
- Client reporting and analytics. Usage, cost, and quality dashboards for your accounts - built and maintained by you, not requested from someone else.
- Feeding fixes back into the product. If you've solved the same class of problem manually more than twice, that's a signal, not a routine. You're expected to flag it and work with product/eng to get it built into tooling instead of staying your recurring task.
What you bring
We don't require a specific number of years managing enterprise accounts. We do require that you can walk us through a time you diagnosed a technical issue for a client yourself - not escalated it, diagnosed and fixed it - and explain your reasoning.- Comfort reading call transcripts and logs, and forming a real hypothesis about why a call failed (ASR vs. LLM vs. TTS vs. telephony) rather than guessing.
- Ability to run a proper root-cause analysis over technical data - correlating timestamps, error codes, and call metadata across systems until you've found the actual point of failure, not just the symptom.
- Comfort working with APIs and webhooks, and enough technical fluency to read a payload, spot a malformed request, or trace where a call in a pipeline went sideways.
- Ability to write and iterate on LLM prompts with precision - this is structured, testable work, not creative writing.
- SQL literacy, or the willingness to get there fast, so you can pull your own numbers instead of waiting on someone else's report.
- Written clarity. Client updates, escalation notes, and prompt specs are all documents. You should enjoy writing them well.
- Comfort operating at the seam between "this is a relationship problem" and "this is an engineering problem" - and good judgment about which is which, and when to pull in someone else.
Nice to have
- Prior exposure to voice AI, contact centers, or CPaaS/telephony infrastructure (SIP, IVR, DLT).
- Experience with BigQuery, Metabase, or similar tools for building your own reporting instead of requesting it.
- Any hands-on prompt engineering or LLM evaluation experience.
- You've had to explain a technical root cause to a non-technical stakeholder without losing them or the deal.
Success in this role looks like
- Repeat escalations for the same class of failure go down for your accounts, because you pushed the fix upstream instead of solving it manually each time.
- SLA and latency breach rates trend down on your accounts, quarter over quarter.
- You walk into renewal and expansion conversations with your own data and a clear point of view - not someone else's slide.
- The recurring manual fix you've owned gets productized into shared tooling instead of remaining a task on your plate.
Why Bolna
- Direct access to founders from day one - you’ll be in the room for decisions most people at your stage never see
- Fastest learning curve available in B2B AI sales, product, and customer success
- Real ownership, real customers, real consequences - not a support function
- Value compensation + meaningful ESOP
- Bengaluru office, in-person team collaboration