Technical Program Manager -— Eka Agents

Technical Program Manager — Eka Agents Bangalore (in-office) | 3–4 years experience About Eka Care Eka Care is India's AI-native Health OS — a single platform bringing together AI-driven EMR for doctors, hospital management tools, ABHA/ABDM integration, and developer APIs to power the full care ecosystem. Built by Orbi Health Pvt Ltd, Eka Care today serves 33,000+ clinics and hospitals, 90,000+ doctors, has created 20Mn+ ABHA health IDs, and manages 140Mn+ health records — with 8Mn+ conversations handled by its MedAssist AI agent and 1Mn+ EkaScribe voice-to-prescription sessions completed. Institutional partners include Apollo Hospitals, Medanta, Samsung, and the National Health Authority. The platform is NHA-approved, FHIR- and ABDM-compliant, and secured on AWS and Google Cloud. Eka's product suite spans EkaDoc EMR (the modern clinic OS), EkaScribe.AI (ambient, voice-to-prescription documentation in 15+ Indian languages), Eka PHR (a patient's health history in one place), a Clinical Decision Support System (real-time, evidence-based guidance for clinicians), and a Developer Portal for building AI-powered health apps. Eka Agents — voice and chat AI for clinical and patient-facing use cases — is part of this same stack. About Eka Agents Eka Agents is Eka Care's conversational AI platform for healthcare. We build agents that run on WhatsApp, voice, and our own voice console — for enterprises around the world. The range is genuinely wide: self-care chatbots for African public-health programs tackling HIV, TB, and maternal & child health; full-stack platforms for Indian enterprises handling appointment booking, insurance, or home care; and clinical decision support for doctors. The surface area is large and growing. Delivering this well isn't a coordination job. It takes someone who understands the product deeply, understands the customer deeply, and can hold both at once. What you'll own You'll run all agentic programs — generally 5–6 enterprise programs in parallel — end to end. Roughly 80% post-sales delivery, 20% pre-sales (solutioning, demos, scoping, and effort estimation alongside the sales conversation). You'll own delivery outcomes that drive commercial success for our clients: renewals, upsell, and margin. The four things this role is actually about: 1. Communication and stakeholder management You are Eka's face to enterprises across the world. That's the job before it's anything else. Language: fluent, confident English — you'll be in rooms with customers well outside India. Hindi and other Indian languages are a plus for Indian accounts; any additional language is a real advantage. Writing: your written craft has to match your spoken craft — SOWs, status reports, RCAs, runbooks, release notes. Most delivery is won or lost in writing. Reading the room: enterprise customers have internal hierarchies and politics. You'll need to work out who decides, who blocks, and who to call and when — building real relationships, not just calendar invites. Making people successful: you know how to get people to back you, and how to back them so they look good internally. Your champion's success is your success. Shielding the team: you absorb churn so engineering doesn't, without the customer ever feeling unheard. Range across cultures: an African public-health program, an Indian hospital chain, and a global buyer expect completely different things. You adjust register without losing substance. Saying no well: "no," "not yet," and "here's what we can do" — delivered so the relationship gets stronger, not weaker. Translating both ways: customer intent into an engineering-ready spec, and technical constraint into something a hospital COO or program director can act on. 2. Deep product and technical understanding AI is a complex domain. Without real understanding, program management becomes a tax on the tech team. This role exists to do the opposite. Know the product cold — every channel, every capability, and the honest edges of each. Be hands-on: write and iterate prompts, build and run evals, test before you deliver, read logs, pull latency and failure signals from metrics, read API docs, and sanity-test APIs yourself. Be fluent with AI tooling generally — you use it in your own work, not just discuss it. Have judgment on claims: know what's a real capability, what's a stretch, and what's a false claim. Know when to say yes, when to hold the line, and when to be smart about sequencing. Never sell something the team can't build. You don't need to write production code — you do need to be a very smart non-engineer who can operate the product properly. 3. Defining and owning success metrics Define success criteria per project at kickoff — not retrofitted after go-live. Own the basics: timelines, milestones, go-live TAT, scope, burn. Own the AI-specific metrics for each project, and pick the right ones for the use case. Report honestly, including when the number is bad. Connect delivery metrics to commercial outcomes. 4. The customer feedback loop Run a continuous loop: capture → triage → into product and engineering → close back with the customer. Separate one-off customer noise from real product signal that should shape the roadmap. Treat the loop as the growth engine — upsell originates here, not in a sales deck. How we work Linear and Notion, used properly, daily. Non-negotiable. Disciplined daily updates — internal and customer reviews. In-office, Bangalore.