Generative AI Lead Engineer

Nexifyr Consulting Pvt Ltd; Hyderabad; On-site; INR 3,800,000 - 7,000,000; Full time

Job summary

Nexifyr Consulting Pvt Ltd is seeking a Generative AI Lead Engineer to architect and deliver GenAI systems for our EdTech platforms.

You will own the GenAI roadmap, oversee multi-agent and edge deployments, and lead a hands-on team to translate early product concepts into scalable, cost-efficient classroom solutions.

Role requires deep GenAI/LLM, MLOps, and Python expertise to build robust software that performs in low-bandwidth environments.

Qualifications

• 8+ years in software or ML engineering with GenAI/LLM production experience.
• Experience leading engineering projects from design to production.
• Hands-on with LLMs, RAG, agents, embeddings, and an agent framework such as Google ADK or LangGraph.
• Strong Python and fundamentals (APIs, testing, containers, cloud, CI/CD).
• Experience with relational and NoSQL databases and vector databases for retrieval.
• Security awareness for systems handling sensitive data.
• Real LLMOps/MLOps experience in production.
• Track record of optimizing AI systems for cost, latency, and reliability at scale.

Responsibilities

• Own GenAI roadmap from prototypes through to production and guide architectural decisions.
• Design multi-agent systems and build production-grade RAG pipelines.
• Lead edge and on-device inference for low-connectivity environments.
• Develop a systematic approach to prompt engineering and versioning.
• Improve cost, latency, and throughput via caching, batching, and routing.
• Establish CI/CD and observability with tracing and cost monitoring.
• Build evaluation harnesses and regression suites (LLM-as-judge, A/B testing).
• Implement guardrails and privacy compliance; ensure secure coding practices.
• Provide technical leadership and set engineering standards for GenAI across teams.

Skills

GenAI/LLM systems

Python

APIs

Testing

Containers

Cloud

CI/CD

RAG

Agents & tool use

Embeddings

Tools

Langfuse

LangSmith

LangGraph

Google ADK

Ollama

vLLM

Job description

About Us:

We are one of India's largest EdTech social enterprises, reaching close to 25
million learners across more than 100,000 schools. Our AI-enabled products make quality
education accessible at scale, spanning personalised adaptive learning, AI teaching assistants,
and intelligent classroom devices, and are designed to operate in low-bandwidth, infrastructure-
constrained environments. As we expand our investment in Generative AI, we are seeking a
senior, hands-on Lead Engineer to design and build our GenAI stack and set the technical
direction for the team.

Role Summary

As the Generative AI Lead Engineer, you will serve as the senior engineer and technical lead
for our GenAI initiatives. You will design the architecture, build the most complex components,
and set the technical direction for the team. The systems you own will range from multi-agent
applications and RAG pipelines to models that run on classroom devices at the edge. This is a
hands-on role focused on building and solving the most challenging technical problems,
translating early-stage product concepts into production GenAI that is scalable and cost-efficient
to operate.

Key Responsibilities

• Technical vision & new development: Own the GenAI roadmap from early prototypes
  through to production, and lead decisions on model selection, build versus buy, and
  architecture.


• Multi-agent & GenAI systems: Design multi-agent systems (orchestration, tool use,
  memory, and agent-to-agent communication via A2A and MCP) and build production-
  grade RAG pipelines.


• Edge AI: Lead edge and on-device inference for low-connectivity environments, using
  quantization and distillation to run open-weight models locally (e.g., Ollama, vLLM).


• Prompt engineering: Build a systematic, versioned, and tested approach to prompt
  development.


• Optimisation: Improve cost, latency, and throughput across our AI systems through
  caching, batching, and model routing.


• Deployment & observability: Establish CI/CD for AI systems and instrument tracing
  and cost, latency, and quality monitoring (e.g., Langfuse, LangSmith).


• Quality & testing: Build evaluation harnesses and regression suites (LLM-as-judge, A/B
  testing) so that every change can be measured.


• Guardrails & Responsible AI: Implement guardrails (validation, safety filters, fallbacks)
  and build in bias and hallucination mitigation and data-privacy compliance.


• Application security: Build security into the GenAI stack, including secure coding
  practices, defences against prompt injection and data leakage, dependency and
  vulnerability scanning, and protection of sensitive user data.


• Technical leadership: Set the technical direction and engineering standards for GenAI,
  maintain engineering quality through design and code reviews, and work closely with
  Product, Design, and QA to keep delivery on track.

Required Skills & Qualifications

• 8+ years in software or ML engineering, including at least 3 years building GenAI/LLM
  systems that have run in production (not just prototypes).


• Experience tech-leading engineering projects from design through to production.


• Deep hands-on experience with LLMs, RAG, agents and tool use, and embeddings,
  plus at least one modern agent framework (e.g., Google ADK, LangGraph).


• Strong prompt-engineering skills and an evaluation-driven mindset.


• Strong Python and solid engineering fundamentals (APIs, testing, containers, cloud,
  CI/CD).


• Solid experience with databases, including relational and NoSQL stores, and vector
  databases for retrieval.


• Familiarity with application security and secure development practices, particularly for
  systems handling sensitive user data.


• Real LLMOps/MLOps experience covering deployment, monitoring, and observability in
  production.


• A track record of optimising AI systems for cost, latency, and reliability at scale.

Preferred Qualifications

• Edge AI and on-device inference experience (quantization, distillation, constrained
  hardware).


• Fine-tuning and PEFT methods such as LoRA.


• Multimodal experience (speech, vision, OCR).


• Full-stack development experience across frontend and backend.


• Experience in EdTech or scalable SaaS platforms.


• Experience deploying and running AI workloads on Google Cloud Platform (GCP).


• Familiarity with responsible-AI and data-privacy frameworks.