Lead Software Engineer- AI Agent Developer

JPMorganChase; Bengaluru; Sur place; INR 4 000 000 - 7 000 000; Plein temps

Résumé du poste

JPMorganChase is seeking a Lead Software Engineer (AI/ML) in Bengaluru to design and operate scalable autonomous agent systems for 100K+ customers. You will lead the development of agentic platforms combining strong software engineering with hands-on expertise in foundation models and LLM-powered architectures.

You will architect planner/executor patterns, tool ecosystems, memory and context management, and observability, while mentoring the team to uphold production-grade standards and

Qualifications

• 8+ years of software engineering experience with large-scale production systems.
• Proven experience building/operating LLM/agentic systems in production with low latency.
• Strong knowledge of foundation models, LLM prompting, tool-augmented generation.
• Experience with knowledge graphs (Neo4J preferred) and OKFs.
• Expertise in context and memory engineering, evaluation methodologies, and quality metrics.
• Experience implementing observability (tracing/metrics/logs, dashboards, alerts).
• Proficiency in Python, Java, Go, or TypeScript; solid API and distributed-systems fundamentals.
• Familiarity with retrieval systems, vector databases, and hybrid search patterns.

Responsabilités

• Architect scalable autonomous agent systems that are production-grade with high reliability and security.
• Design agentic architectures including planner/executor patterns and multi-agent orchestration.
• Build robust tool interfaces and ecosystems (APIs, function calling, permissions, safe execution).
• Engineer context and memory strategies (short/long-term memory, summarization, grounding, caching).
• Own model evaluation and quality measurement (offline/online evals, regression testing).
• Implement end-to-end observability (tracing, metrics, logging, telemetry).
• Drive cost and performance optimization across models and infrastructure (throughput, batching, caching, routing).
• Mentor engineers and establish production AI engineering standards.

Connaissances

8+ years experience

LLM/agentic systems

foundation models

context & memory engineering

observability

Python

Java

Go

TypeScript

APIs & distributed systems

retrieval systems

Description du poste

Job Description:

Join us to advance your software engineering career while building impactful technology solutions. Grow your skills and make a difference with a collaborative team.

We’re seeking a Lead Software Engineer (AI/ML) to build and operate scalable, low-latency autonomous agent systems in production. You’ll lead the design and delivery of agentic platforms that reliably serve 100K+ customers, combining strong software engineering fundamentals with deep hands‑on expertise in foundation models and LLM-powered architectures.

Job responsibilities

• Architect and ship autonomous agent systems that are scalable, resilient, and production-grade (latency, reliability, security, and uptime).
• Design agentic architectures (planner/executor patterns, multi-agent orchestration, tool use, retrieval, and workflow execution).
• Build robust tool interfaces and tool ecosystems (APIs, function calling patterns, permissions, and safe execution).
• Engineer context, memory, and state management strategies (short/long-term memory, summarization, retrieval, grounding, and caching).
• Own model evaluation and quality measurement (offline/online evals, regression testing, and failure-mode analysis).
• Implement end-to-end observability for LLM systems (tracing, metrics, logging, prompt/tool telemetry, incident response).
• Drive cost and performance optimization across models and infrastructure (throughput, batching, caching, routing, quantization, and inference efficiency).
• Mentor engineers and set engineering standards for production AI systems (code quality, design reviews, and operational excellence).

Required qualifications, capabilities and skills

• 8+ years of software engineering experience, including leading design and delivery of large-scale production systems.
• Demonstrated experience building and operating LLM/agentic systems in production with strict latency and reliability requirements.
• Strong understanding of foundation models, LLM behavior, prompting strategies, and tool-augmented generation patterns.
• Strong understanding of Knowledge graph (Neo4J preferred), Open Knowledge formats (OKFs)
• Proven expertise in context and memory engineering, model evaluation methodologies, and quality measurement.
• Experience implementing observability for distributed systems and AI applications (tracing/metrics/logs, dashboards, alerting, on-call readiness).
• Strong programming skills in one or more of: Python, Java, Go, or TypeScript; plus strong API and distributed-systems fundamentals.
• Familiarity with retrieval systems, vector databases, and hybrid search patterns.

Preferred qualifications, capabilities and skills

• Agentic AI architecture: planner/executor patterns, multi-agent orchestration, tool-use design, retrieval workflows
• LLM/foundation model expertise: prompting strategies, tool-augmented generation, model behavior understanding
• Context & memory engineering: short/long-term memory, summarization, retrieval grounding, caching strategies
• Knowledge graphs: Neo4J (preferred), Open Knowledge Formats (OKFs)
• Retrieval systems: vector databases, hybrid search patterns