Full Stack AI Engineer

About the Role

We are looking for a Full Stack AI Engineer who can take an ambiguous problem and turn it into a complete, production-ready AI product.

This is a builder role.

You will work across the entire stack — AI models, agents, backend services, APIs, databases, data pipelines, frontend applications, infrastructure, and production deployment. You should be comfortable deciding what needs to be built, writing the code, deploying it, measuring whether it works, and continuously improving it.

We are not looking for someone who only builds notebooks, trains models, writes prompts, or creates architecture diagrams for another team to implement. We want engineers who ship complete products.

A typical project might involve designing an agentic workflow, building a retrieval pipeline, writing Python APIs, creating a React interface, integrating enterprise data, deploying to Kubernetes, implementing evaluations, and debugging the application in production.

The distance between an idea and working software should be measured in weeks, not quarters.

This role is based in Hyderabad and is 6 days per week in the office.

We are looking for:

1. A person that can smoothly navigate extreme ambiguity
2. Full stack software builder
3. AI depth, must have built a production AI system
4. Excellent top-notch communication and stakeholder management
5. Prioritizes growth over work-life balance

What You'll Do

Build AI Products End-to-End

• Own AI applications from problem definition through architecture, development, deployment, and production operation.
• Translate ambiguous product and business requirements into working software.
• Build across AI/ML, backend services, APIs, databases, frontend interfaces, data pipelines, authentication, infrastructure, and observability.
• Rapidly prototype, test with real users and data, and turn successful ideas into production-grade systems.
• Make pragmatic engineering decisions based on speed, reliability, simplicity, maintainability, and user value.

AI, LLMs & Agents

• Build production applications using commercial and open-source foundation models.
• Design RAG systems, agentic workflows, tool/function calling, structured outputs, memory, human-in-the-loop workflows, and multi-agent systems where appropriate.
• Work with frameworks such as LangGraph, LangChain, Semantic Kernel, LlamaIndex, or equivalent tools.
• Build retrieval systems using embeddings, vector search, BM25, hybrid retrieval, reranking, metadata filtering, and knowledge graphs.
• Design prompt and context engineering strategies for complex workflows.
• Evaluate model choices based on accuracy, latency, reliability, security, and cost.
• Build automated evaluations and regression tests for AI behavior.
• Fine-tune or adapt models when prompting and retrieval are insufficient.

Backend, Frontend & Data

• Build production backend systems primarily in Python using FastAPI, Flask, Django, or similar frameworks.
• Design APIs, asynchronous workflows, background jobs, queues, caching layers, and event-driven systems.
• Work with PostgreSQL, SQL Server, MongoDB, Redis, Snowflake, and other production data stores.
• Build modern applications using React, Next.js, TypeScript, JavaScript, or equivalent frameworks.
• Create interfaces for copilots, conversational AI, workflow automation, analytics, review queues, and operational applications.
• Implement streaming responses, real-time updates, authentication, permissions, and API integrations.
• Build ingestion and transformation pipelines for structured and unstructured enterprise data.
• Work with documents, databases, APIs, event streams, images, logs, and operational datasets.
• Maintain provenance, permissions, metadata, and traceability across enterprise information.

ML & Computer Vision

• Use classical ML or deep learning when it is better suited to the problem than an LLM.
• Build systems involving classification, forecasting, anomaly detection, ranking, recommendations, optimization, or prediction.
• Build computer-vision applications involving detection, classification, segmentation, OCR, tracking, or image/video analysis.
• Work with PyTorch, TensorFlow, Hugging Face, OpenCV, or equivalent tools.
• Understand model development, evaluation, inference, and productionization.

Deploy & Operate What You Build

• Deploy applications across AWS, Azure, GCP, on-premises, hybrid, or edge environments.
• Containerize and operate applications using Docker and Kubernetes.
• Build CI/CD pipelines, automated testing, monitoring, and observability.
• Own reliability, latency, availability, security, evaluation, cost, and scalability.
• Debug failures across application code, AI models, data, infrastructure, and integrations.
• Build retries, fallbacks, rollback mechanisms, and human intervention into critical systems.

Integrate With Enterprise Systems

• Connect AI applications to enterprise platforms, databases, APIs, and operational systems.
• Integrate with systems such as ERP, MES, PLM, CRM, data warehouses, IoT platforms, document repositories, and legacy applications.
• Work within enterprise networking, security, and data-governance constraints.
• Implement authentication, authorization, secrets management, auditability, permissions, and data isolation.
• Build AI systems capable of safely operating on sensitive enterprise data.

What You Bring

• 5+ years of software engineering experience, with meaningful experience building AI/ML-powered products. Exceptional candidates with less experience but strong demonstrated ability will be considered.
• Strong hands-on programming ability in Python.
• Experience building complete production applications rather than isolated models, notebooks, or proofs of concept.
• Strong backend fundamentals including APIs, databases, distributed systems, and application architecture.
• Experience with React, Next.js, TypeScript, JavaScript, or equivalent frontend technologies.
• Strong understanding of LLMs, RAG, agents, tool calling, embeddings, vector search, prompt/context engineering, AI evaluation, and ML fundamentals.
• Experience with SQL and production databases.
• Experience with at least one major cloud platform: AWS, Azure, or GCP.
• Experience with Docker, Kubernetes, or equivalent production infrastructure.
• Understanding of production AI concerns including reliability, latency, security, observability, evaluation, and cost.
• Strong debugging skills across the full stack.
• Ability to independently turn loosely defined requirements into working software.
• Strong product judgment, high agency, technical curiosity, and a bias toward shipping.
• Bachelor's degree in Computer Science, Engineering, Data Science, or a related field, or equivalent demonstrated experience.

Preferred / Top-Candidate Signals

• You've independently shipped an AI application from database → backend → AI → frontend → production.
• You've built production RAG or agentic systems, not just demos.
• You've used LangGraph, LangChain, Semantic Kernel, LlamaIndex, or similar frameworks.
• You understand when not to use an LLM or agent.
• You've worked with hybrid retrieval, reranking, vector search, or knowledge graphs.
• You've built with both commercial and open-source models.
• You've deployed ML or computer-vision systems into production.
• You have experience with React/Next.js + Python/FastAPI or a comparable modern stack.
• You have experience with Kubernetes, cloud infrastructure, and production observability.
• You have integrated software with complex enterprise or industrial systems.
• Experience in manufacturing, industrial, supply chain, logistics, engineering, energy, aerospace, automotive, or other physical-world environments is a strong plus.
• You've built meaningful side projects, open-source software, startups, or substantial systems outside your assigned responsibilities.
• You have a history of turning vague ideas into shipped products.

What Makes Someone Exceptional

The strongest engineers in this role combine three abilities:

AI Engineering — Choose the right model, retrieval approach, agent architecture, evaluation method, or ML technique.

Software Engineering — Build everything around the intelligence: frontend, backend, data, APIs, infrastructure, security, integrations, and deployment.

Product Judgment — Understand what actually needs to be built and rapidly turn it into something users can use.

A typical week might involve designing an agent workflow, writing FastAPI services, building a React interface, creating a retrieval pipeline, connecting enterprise data, deploying to Kubernetes, implementing evaluations, and debugging real-world behavior.

You should not need five different teams to turn an idea into a working product. You should be able to build.

Why This Role

• Build entire products, not isolated models or prototypes.
• Own the full stack: AI, backend, frontend, data, infrastructure, and deployment.
• Ship quickly: move from idea to working software in weeks.
• Work across modern AI: LLMs, agents, RAG, ML, computer vision, optimization, and enterprise data.
• Solve real-world problems: build software used in complex operational environments.
• See your work in production: own the path from first commit to real users.
• Compensation is flexible for exceptional candidates.