Manager - AI Accelerator Program

National%20institute%20for%20smart%20government; Khordha; On-site; INR 1,200,000 - 1,800,000; Full time

Job summary

National Institute for Smart Government seeks a Forward Deployed AI / Software Engineer to work with enterprise clients, translating business requirements into AI-powered solutions. You will design, build, and deploy production-grade software, focusing on agents, LLM prompts, and robust workflows across APIs, databases, and cloud platforms.

Remote CST-hours friendly role with exposure to end-to-end development, from discovery to production, across cross-functional teams.

Qualifications

• Strong professional experience building and deploying production-grade software.
• Strong proficiency in Python, TypeScript, or a similar programming language.
• Hands-on experience building AI-powered applications or agentic systems.
• Practical understanding of LLMs, prompts, context management, tool use, memory, and model selection.
• Experience building or working with AI agents/agentloops involving planning, tool use, execution, observation, and recovery.
• Understanding of workflow orchestration, graphs, state machines, or multi-step AI workflows.
• Experience working with APIs, databases, cloud infrastructure, and modern software engineering practices.
• Experience taking a problem from discovery to prototype to validation to production.
• Ability to work directly with clients or business stakeholders and translate business requirements into technical solutions.
• Strong debugging, problem-solving, communication, and ownership skills.
• Comfortable working remotely with teams operating in CST business hours.

Responsibilities

• Work directly with clients to understand business requirements, workflows, technical environments, and operational challenges.
• Translate ambiguous business problems into technical requirements, solution designs, prototypes, and production-ready applications
• Build AI-powered applications, agentic workflows, APIs, data pipelines, and full-stack solutions
• Design and implement systems around AI models, including tools, context, memory, permissions, runtime controls, observability, and feedback mechanisms.
• Build and refine agent workflows covering planning, tool selection, execution, observation, recovery, and termination.
• Develop workflow orchestration using graphs, state machines, and multi-step execution patterns
• Build evaluation mechanisms to measure AI application quality, reliability, task completion, latency, and cost.
• Integrate AI solutions with enterprise applications, APIs, databases, cloud platforms, and third-party services.
• Troubleshoot AI-system issues such as hallucinations, context loss, unreliable tool calls, looping, and prompt-injection risks.
• Collaborate with product, engineering, data, security, and client teams to take solutions from discovery to production.
• Clearly communicate technical decisions, risks,limitations, and trade-offs to technical and non-technical stakeholders.

Skills

Python

TypeScript

AI applications

Agentic systems

LLMs prompts

Workflow orchestration

APIs

Cloud services

Production deployment

Client collaboration

Debugging skills

Tools

Docker

Kubernetes

CI/CD

Infrastructure as Code

AWS

Azure

Google Cloud

Infrastructure as Code

Job description

Job Summary

Position: Forward Deployed AI / Software Engineer

Responsibilities

• Work directly with clients to understand business requirements, workflows, technical environments, and operational challenges.
• Translate ambiguous business problems into technical requirements, solution designs, prototypes, and production-ready applications
• Build AI-powered applications, agentic workflows, APIs, data pipelines, and full-stack solutions
• Design and implement systems around AI models, including tools, context, memory, permissions, runtime controls, observability, and feedback mechanisms.
• Build and refine agent workflows covering planning, tool selection, execution, observation, recovery, and termination.
• Develop workflow orchestration using graphs, state machines, and multi-step execution patterns
• Build evaluation mechanisms to measure AI application quality, reliability, task completion, latency, and cost.
• Integrate AI solutions with enterprise applications, APIs, databases, cloud platforms, and third-party services.
• Troubleshoot AI-system issues such as hallucinations, context loss, unreliable tool calls, looping, and prompt-injection risks.
• Collaborate with product, engineering, data, security, and client teams to take solutions from discovery to production.
• Clearly communicate technical decisions, risks,limitations, and trade-offs to technical and non-technical stakeholders.

Required Skills

• Strong professional experience building and deploying production-grade software.
• Strong proficiency in Python, TypeScript, or a similar programming language.
• Hands-on experience building AI-powered applications or agentic systems.
• Practical understanding of LLMs, prompts, context management, tool use, structured outputs, memory, and model selection.
• Experience building or working with AI agents/agentloops involving planning, tool use, execution, observation, and recovery.
• Understanding of workflow orchestration, graphs, state machines, or multi-step AI workflows.
• Experience working with APIs, databases, cloud infrastructure, and modern software engineering practices.
• Experience taking a problem from discovery to prototype to validation to production.
• Ability to work directly with clients or business stakeholders and translate business requirements into technical solutions.
• Strong debugging, problem-solving, communication, and ownership skills.
• Comfortable working remotely with teams operating in CST business hours.

Preferred (Bonus) Skills

• Experience with AI agent frameworks, workflow engines, coding agents, or custom AI orchestration systems.
• Experience with RAG, embeddings, vector databases/search, knowledge graphs, fine-tuning, or model routing.
• Experience with AI evaluation, regression testing, human evaluation, or production monitoring.
• Familiarity with AI observability, tracing, token/cost analysis, latency monitoring, and quality dashboards.
• Experience building multi-agent or multi-model systems.
• Understanding of AI security risks including prompt injection, data leakage, excessive agency, and insecure tool usage.
• Experience with AWS, Azure, or Google Cloud.
• Experience with Docker, Kubernetes, Infrastructure as Code, and CI/CD.
• Experience in consulting, solutions engineering, professional services, or other customer-facing technical roles.
• Experience working with complex enterprise environments.