Backend Solution Engineer - SDE II

Eka.Care; Bengaluru; On-site; INR 1,500,000 - 2,200,000; Full time

Benefits offered by this job

Medical Insurance

Parental Support

Retirement Benefits

Salary Advance Policy

Job summary

Eka Care, a health AI company, is seeking aBackend/Systems engineer to build production-grade systems integrating AI, streaming audio, and LLM-based tooling. You will own features from design to production and on-call, delivering reliable services at scale.

You will collaborate with product and clinical teams to turn complex clinical problems into robust, scalable infrastructure, ensuring low latency, high availability, and strong monitoring across the EkaScribe stack.

Qualifications

• 4+ years building and running production backend systems.
• Strong system design fundamentals: distributed systems, concurrency, consistency trade-offs, API design and data modelling.
• Strong exposure to system reliability principles.
• High-speed delivery capability in a startup environment.
• Hands-on with Python, FastAPI, MCPs, Design Principles, AWS and Kubernetes: deploying, autoscaling and debugging services in production.
• Solid experience with Redis and Postgres: event-driven design, caching, and schema/query design at scale.
• Hands-on LLM application and agent experience in production: prompting, tool calling, structured outputs, RAG, and orchestration.
• Experience building evals and observability for LLM systems: tracing, scoring, and regression gates.
• A strong ownership mindset; you own it in production, not just on merge.

Responsibilities

• Architect and build on EkaScribe's pipeline: streaming audio intake, ASR, transcript processing, LLM-based document generation, and doctor edit-and-publish loop.
• Own reliability end to end: SLOs, fallbacks, retries, idempotency, backpressure, incident response and postmortems.
• Build evals and observability for LLM systems: tests, signals, tracing, and evidence-based improvements.
• Build agentic systems: tool-using agents, multi-step orchestration, memory, retrieval, and guardrails.
• Design for scale and optimized solutions: queue-based async processing, caching, batching, token/GPU budgets, multi-tenant isolation.
• Ship at a high pace: short iterations, safe deploys, feature flags, and clear design docs.
• Work closely with product, clinical and mobile teams to translate clinical problems into scoped systems.

Skills

System design

Distributed systems

Reliability

Delivery speed

Python

FastAPI

AWS

Kubernetes

Redis

Postgres

LLM apps

Observability

Tools

Python

FastAPI

MCPs

AWS

Kubernetes

Redis

Postgres

LangGraph

CrewAI

Job description

Eka Care is a health AI company on a mission to make world-class clinical intelligence available to every doctor, patient, and healthcare enterprise.

We build the models, data infrastructure, and clinical workflows that turn everyday care — a consultation, a prescription, a lab report — into structured, interoperable health data, so that better decisions can be made at the point of care and across the enterprise, whether that is a hospital, a diagnostic lab, or an insurer. Our products — EkaScribe, EkaDoc, EkaAgents, EkaAPI's and EkaPHR — cover the full span of clinical, patient, and enterprise use cases, and plug into national digital health infrastructure such as India's Ayushman Bharat Digital Mission in India.

The role

We're looking for a strong backend and systems engineer who has hands-on exposure with AI. Engineering comes first and AI second. You'll design, build and own production systems that combine core engineering, streaming audio, LLMs and agents, and you'll run them at scale. You'll own features from design doc to production to on-call, and you'll ship fast without breaking things.

What you'll do

• Architect and build on EkaScribe's pipeline: streaming audio intake, ASR, transcript processing, LLM-based document generation, and the doctor's edit-and-publish loop. Low latency and high availability are requirements.
• Own reliability end to end: SLOs, fallbacks across model providers, retries, idempotency, backpressure, graceful degradation, incident response and postmortems.
• Build evals and observability for LLM systems: offline eval suites, online quality signals, hallucination and omission tracking, and LLM tracing. Every prompt or model change should ship with evidence that it's better.
• Build agentic systems: tool-using agents, multi-step orchestration, memory and retrieval, and guardrails that keep agents grounded in what was actually said and recorded.
• Design for scale and optimised solutions: queue-based async processing, caching, batching, token and GPU cost budgets, and multi-tenant isolation.
• Ship at a high pace: short iteration cycles, small safe deploys, feature flags, and clear written design docs.
• Work closely with product, clinical and mobile teams, and turn unclear clinical problems into well-scoped systems.

Must-have

• 4+ years building and running production backend systems , with a track record of owning services others depend on.
• Strong system design fundamentals: distributed systems, concurrency, consistency trade-offs, API design and data modelling.
• Strong exposure and adherence to system reliability principles
• High-speed delivery capability/ startup environment without compromising on reliability
• Hands-on with Python, Fast API, MCPs, Design Principles, AWS and Kubernetes : deploying, autoscaling and debugging services in production.
• Solid experience with Redis and Postgres : event-driven design, caching, and schema and query design at scale.
• Hands-on LLM application and agent experience in production: prompting, tool calling, structured outputs, RAG, and orchestration (LangGraph, CrewAI or custom).
• Experience building evals and observability for LLM systems: test sets, automated and LLM-as-judge scoring, tracing, and regression gates.
• A strong ownership mindset . You own it in production, not just on merge.

Nice-to-have

• Working knowledge of speech/ASR pipelines : streaming audio, ASR model trade-offs, diarization, and handling noisy, multilingual audio.
• Other languages / technology used for high-throughput services
• Vector search (pgvector or similar) and memory systems for agents
• Healthcare or regulated-data experience (ABDM, HIPAA, PII handling)
• Open-source contributions or public writing

Full-Time Employee Benefits:

Insurance Benefits - Medical Insurance, Accidental Insurance

Parental Support - Maternity Benefit, Paternity Benefit Program

Retirement Benefits - Employee PF Contribution, Gratuity, NPS, Leave Encashment

Other Benefits - Salary Advance Policy