Indiranagar, Bengaluru
Location: Indiranagar, Bangalore
Type: Full-time | 5 Days In-Office
About Triplespeed
We incubate viral consumer brands for the US and EU markets and scale them until they become independent companies.
In the last 14 months, we took three apps, across three unrelated verticals, from zero to over $70M in combined annual run rate. We did this bootstrapped, with no outside capital. We're one of the fastest-growing bootstrapped companies in the world.
We’re looking for an Applied AI Engineer to build intelligent, agentic systems that can reason, use tools, manage context, and execute complex tasks reliably. You’ll work at the intersection of LLMs, agent infrastructure, and applied AI engineering to turn emerging AI capabilities into production-ready systems.
What You’ll Do
AI Harnesses: Build the infrastructure and execution environments that enable AI agents to perform complex tasks reliably.
Context Management: Design strategies for managing context windows, maintaining relevant information, and handling long-running agent sessions.
Memory & Retrieval: Develop memory systems that retrieve, store, and surface relevant information across interactions and tasks.
Evaluations (Evals): Build evaluation frameworks to measure agent quality, reliability, and performance. Create test suites to catch regressions and improve outcomes.
Tool Design & MCP: Design robust tool interfaces, schemas, and integrations. Work with APIs, function calling, and the Model Context Protocol (MCP).
Long-Running & Human-in-the-Loop Agents: Build agents capable of executing multi-step tasks over extended periods, with checkpoints, approvals, and human intervention when needed.
Skills & Prompt Engineering: Develop reusable agent skills, system prompts, and instructions that improve task execution and consistency.
State Management: Design mechanisms for maintaining state across agent steps, sessions, and interrupted workflows.
Failure Handling: Implement retries, recovery mechanisms, fallbacks, and safeguards to handle unexpected behavior and tool failures.
Observability: Build logging, tracing, and monitoring systems to understand agent behavior, debug failures, and identify performance bottlenecks.
What We’re Looking For
Strong software engineering fundamentals and experience building production-grade applications.
Hands-on experience building LLM-powered applications, AI agents, or agentic workflows.
Strong programming skills in Python or TypeScript.
Experience working with LLM APIs and understanding their capabilities and limitations.
Familiarity with tool calling, structured outputs, API integrations, and workflow orchestration.
Understanding of stateful systems, asynchronous execution, and failure recovery.
Ability to evaluate AI systems systematically rather than relying on subjective assessments.
Strong debugging skills and an ability to work through ambiguous, technically challenging problems.
A bias toward experimentation, rapid iteration, and shipping production-ready solutions.
Bonus Points
Experience with agent frameworks such as LangGraph, LangChain, or similar technologies.
Familiarity with MCP servers, tool schemas, and agent-to-tool integrations.
Experience building persistent memory and retrieval systems.
Experience with distributed systems, background jobs, queues, and durable workflow execution.
Familiarity with LLM observability and evaluation tools.
Experience building agents that operate autonomously while supporting human approvals and intervention.
What Success Looks Like
Agents can execute complex, multi-step tasks with minimal unnecessary human intervention.
Agent workflows are reliable, observable, and recover gracefully from failures.
Context and memory systems help agents maintain continuity across long-running tasks.
Evals provide measurable signals for improving quality and preventing regressions.
New tools, capabilities, and agent skills can be added without rebuilding the entire system.