About the role
Build production agentic workflows and enterprise AI integrations that real businesses use.
Openings: 2
Experience: 2 to 5 years
Location: Mumbai, Pune or Noida preferred. Strong candidates in Bengaluru, Hyderabad and Chennai also considered. Remote possible for exceptional profiles.
Hiring timeline: Priority hiring. Immediate joiners preferred.
About Our Client
Our client is an applied-AI and product engineering studio that builds production-grade technology with ambitious businesses. Engineering, AI, product thinking and delivery ownership sit in the same team. This role is part of an enterprise AI delivery program where clean engineering, clear communication and measurable results count.
Selected candidates will be on our client's payroll and deployed on their enterprise customer engagements as part of their own delivery teams.
The Role
You will design and ship real AI applications inside a four-person forward-deployed pod. We need an engineer who is comfortable talking to business users, writes reliable Python, debugs live workflows and takes features from discovery to production in short iterations.
What You Will Do
• Build agentic processes and RAG workflows that solve concrete business or developer productivity problems.
• Build and debug Python services, APIs, data-retrieval components and orchestration with LangGraph and LangChain.
• Instrument traces, latency, quality signals and failures in Langfuse. Troubleshoot agents when they behave unexpectedly.
• Build and run cloud components on GCP. Integrate enterprise AI tools, Google ADK and API gateways where they fit.
• Apply prompt guardrails, access boundaries, security basics and safe handling of retrieved context.
• Work directly with business and IT stakeholders, clarify fuzzy problems and demo progress regularly.
• Review code, help junior engineers troubleshoot and raise technical risks to the principal engineer.
• Help the team adopt AI in day-to-day development with tools like Cursor and Codex.
What You Bring
• 3 to 5 years of software engineering, with practical delivery of AI-enabled applications.
• Strong Python coding and debugging. You can solve problems without code assistants.
• Applied experience or solid project depth in GCP, LangGraph, LangChain and Langfuse.
• Good grasp of agentic AI, RAG, embeddings, vector databases and AI governance.
• Confident spoken communication with cross-functional stakeholders.
Technical Essentials
• Strong DSA in Python, demonstrated in a live coding round with no AI tools.
• Write, read and debug Python independently, demonstrated in the same live round.
• Competent use of Google Cloud Platform (GCP) for enterprise AI applications.
• LangGraph, LangChain and Langfuse: you can explain where each fits in an agentic application.
• Agentic workflows, retrieval-augmented generation (RAG), embeddings and vectorization, orchestration.
• Enterprise AI components: Google Agent Development Kit (ADK), API gateways, integration patterns.
• Clear communication with business and IT teams. You ask good discovery questions and explain trade-offs in plain English.
• AI governance: prompt guardrails, model security, contextual and data poisoning, safety controls.
Good to Have
• Google Model Armor or similar AI security controls (an advantage, not mandatory).
• LLM evaluation, observability, prompt and version management, tracing, incident debugging.
• Secure integrations, auth, API lifecycle management, production monitoring.
• Using Cursor, Codex or similar tools to raise developer productivity.
• Exposure to underwriting, conversational reporting/MIS or sales and workflow automation.
What Success Looks Like
• Reliable Python-based agentic and RAG features shipped in small, production-ready increments.
• Observable workflows with traceable outputs and documented security and failure handling.
• Smooth handoffs and clear business-facing demos, without depending on long requirement documents.
Selection Process
1. Recruiter screening: communication, role fit, location, notice period, motivation and proof of real engineering ownership.
2. Technical screen: Python coding and debugging without AI tools, plus a practical check on GCP, agentic AI, RAG and governance.
3. Conversation with the engagement lead.
4. Technical interview with the end client, which may include live coding without AI assistance.
All live rounds are on a laptop in a stable setup. You will be briefed on the format in advance.
Why This Role
• Real business and technology problems, worked on with a multidisciplinary delivery team.
• Direct access to stakeholders and ownership of outcomes, not just tasks.
• Hands-on experience in enterprise-scale applied AI and transformation programs.
Openings: 2
Experience: 2 to 5 years
Location: Mumbai, Pune or Noida preferred. Strong candidates in Bengaluru, Hyderabad and Chennai also considered. Remote possible for exceptional profiles.
Hiring timeline: Priority hiring. Immediate joiners preferred.
About Our Client
Our client is an applied-AI and product engineering studio that builds production-grade technology with ambitious businesses. Engineering, AI, product thinking and delivery ownership sit in the same team. This role is part of an enterprise AI delivery program where clean engineering, clear communication and measurable results count.
Selected candidates will be on our client's payroll and deployed on their enterprise customer engagements as part of their own delivery teams.
The Role
You will design and ship real AI applications inside a four-person forward-deployed pod. We need an engineer who is comfortable talking to business users, writes reliable Python, debugs live workflows and takes features from discovery to production in short iterations.
What You Will Do
• Build agentic processes and RAG workflows that solve concrete business or developer productivity problems.
• Build and debug Python services, APIs, data-retrieval components and orchestration with LangGraph and LangChain.
• Instrument traces, latency, quality signals and failures in Langfuse. Troubleshoot agents when they behave unexpectedly.
• Build and run cloud components on GCP. Integrate enterprise AI tools, Google ADK and API gateways where they fit.
• Apply prompt guardrails, access boundaries, security basics and safe handling of retrieved context.
• Work directly with business and IT stakeholders, clarify fuzzy problems and demo progress regularly.
• Review code, help junior engineers troubleshoot and raise technical risks to the principal engineer.
• Help the team adopt AI in day-to-day development with tools like Cursor and Codex.
What You Bring
• 3 to 5 years of software engineering, with practical delivery of AI-enabled applications.
• Strong Python coding and debugging. You can solve problems without code assistants.
• Applied experience or solid project depth in GCP, LangGraph, LangChain and Langfuse.
• Good grasp of agentic AI, RAG, embeddings, vector databases and AI governance.
• Confident spoken communication with cross-functional stakeholders.
Technical Essentials
• Strong DSA in Python, demonstrated in a live coding round with no AI tools.
• Write, read and debug Python independently, demonstrated in the same live round.
• Competent use of Google Cloud Platform (GCP) for enterprise AI applications.
• LangGraph, LangChain and Langfuse: you can explain where each fits in an agentic application.
• Agentic workflows, retrieval-augmented generation (RAG), embeddings and vectorization, orchestration.
• Enterprise AI components: Google Agent Development Kit (ADK), API gateways, integration patterns.
• Clear communication with business and IT teams. You ask good discovery questions and explain trade-offs in plain English.
• AI governance: prompt guardrails, model security, contextual and data poisoning, safety controls.
Good to Have
• Google Model Armor or similar AI security controls (an advantage, not mandatory).
• LLM evaluation, observability, prompt and version management, tracing, incident debugging.
• Secure integrations, auth, API lifecycle management, production monitoring.
• Using Cursor, Codex or similar tools to raise developer productivity.
• Exposure to underwriting, conversational reporting/MIS or sales and workflow automation.
What Success Looks Like
• Reliable Python-based agentic and RAG features shipped in small, production-ready increments.
• Observable workflows with traceable outputs and documented security and failure handling.
• Smooth handoffs and clear business-facing demos, without depending on long requirement documents.
Selection Process
1. Recruiter screening: communication, role fit, location, notice period, motivation and proof of real engineering ownership.
2. Technical screen: Python coding and debugging without AI tools, plus a practical check on GCP, agentic AI, RAG and governance.
3. Conversation with the engagement lead.
4. Technical interview with the end client, which may include live coding without AI assistance.
All live rounds are on a laptop in a stable setup. You will be briefed on the format in advance.
Why This Role
• Real business and technology problems, worked on with a multidisciplinary delivery team.
• Direct access to stakeholders and ownership of outcomes, not just tasks.
• Hands-on experience in enterprise-scale applied AI and transformation programs.