Skip to content
Search jobs inside your AI agent. Claude, ChatGPT, Claude Code, Codex, Gemini, Cursor and Grok
Connect

ChiStats

NewBe an early applicant

AI Agent Engineer

Posted Updated

Description

About ChiStats:

ChiStats is an award-winning Data Science and AI Engineering company building intelligent technology solutions for the Financial Services and BFSI ecosystem. From AI-powered compliance platforms to scalable data products, we help financial institutions simplify complexity, automate operations, and stay ahead in a rapidly evolving regulatory landscape.


Our flagship platform, ComplianceAssist, enables Mutual Funds, AMCs, PMS, and AIFs to transform regulatory compliance into a streamlined, automated, and insight-driven process. As we continue to expand, we are building the next generation of AI-led products and solutions for the WealthTech industry. We are looking for engineers to build applications and personal-agent experiences on top of frontier AI lab platforms, including personal agent products such as Muse, Dot and similar.


You will design agentic workflows, build AI agents that use tools and APIs to get real work done, and integrate apps through platform SDKs so agents can act on users' behalf across services.

Work Time: 2:30 PM - 11:30 PM

Key responsibilities:

- Design, build and ship AI agents and personal-agent use cases on frontier lab platforms (e.g., Muse, Dot and comparable ecosystems).

- Build multi-step agentic workflows: planning, tool selection, task decomposition, memory, and human-in-the-loop checkpoints.

- Integrate third-party apps and services through Apps SDKs, tool/function-calling interfaces and REST/GraphQL APIs.

- Build and maintain tool connectors (including MCP servers or similar tool protocols) that expose data and actions to agents securely.

- Implement context and memory management, including retrieval (RAG), session state and user personalization.

- Engineer and iterate on prompts, system instructions and tool schemas to improve reliability, accuracy and cost.

- Set up evaluation, testing and observability for agent behaviour (success rates, failure analysis, latency and token cost).

- Handle authentication and permissions for connected apps (OAuth 2.0, API keys, scoped access) with attention to user privacy and safety.

- Collaborate with product and design to turn user needs into agent capabilities, and prototype quickly to validate ideas.

- Keep up with fast-moving frontier lab releases, SDKs and agent frameworks, and bring relevant improvements into the product.

Required Skills:

1. Agentic AI and LLMs:

- Hands-on experience building AI agents and agentic workflows (tool use, planning, multi-step execution, error recovery).

- Working knowledge of LLM APIs from frontier labs (e.g., Anthropic, OpenAI, Google) including function/tool calling, structured outputs and streaming.

- Experience with at least one agent framework or SDK (e.g., Claude Agent SDK, OpenAI Agents SDK / Apps SDK, LangChain/LangGraph, LlamaIndex, CrewAI, AutoGen).

- Prompt engineering and context engineering for reliable agent behaviour.

- Understanding of RAG, embeddings and vector databases (e.g., Pinecone, Weaviate, pgvector, Chroma).

2. Integrations and APIs:

- Strong experience integrating apps via SDKs, webhooks and REST/GraphQL APIs.

- Familiarity with tool protocols such as the Model Context Protocol (MCP) or plugin/connector architectures.

- Authentication and authorization: OAuth 2.0, token handling, scoped permissions.

3. Software engineering:

- Proficiency in Python and/or TypeScript/Node.js.

- Building backend services and APIs (FastAPI, Flask, Express or similar).

- Async programming, background jobs and event-driven design for long-running agent tasks.

- Git, CI/CD, automated testing and code review practices.

- Cloud platforms (AWS, GCP or Azure), Docker and basic deployment/DevOps.

4. Soft skills:

- Product mindset: thinking in terms of user use cases, not just models.

- Comfort with ambiguity and rapid prototyping in a fast-changing space.

- Clear written and verbal communication; able to document agent designs and trade offs.

Good to have:

- Prior work on personal assistant or companion apps (e.g., Muse, Dot or similar consumer AI products).

- Experience with multi-agent systems and agent-to-agent orchestration.

- Computer-use or browser-automation agents (e.g., Playwright, Puppeteer).

- LLM evaluation and observability tooling (e.g., LangSmith, Langfuse, Braintrust, Arize).

- Long-term memory and personalization systems for agents.

- Mobile or frontend experience (React, React Native, Swift/Kotlin) for agent-facing UIs.

- Knowledge of AI safety, guardrails, prompt-injection defence and data privacy practices.

- Open-source contributions or public projects in the agentic AI space.

Requirements:

- 2 - 6 years of professional software engineering experience, with at least 1 year working hands-on with LLMs or AI agents.

- Bachelor's degree in Computer Science, Engineering or a related field, or equivalent practical experience.

- A portfolio, GitHub or demonstrable projects showing agents or LLM integrations built end to end.

Skills Summary:

- Core: AI Agents, Agentic Workflows, LLMs, Tool/Function Calling, Personal Agents.

- Platforms: Frontier Lab Apps, Muse, Dot, Anthropic, OpenAI and Google APIs.

- SDKs & Frameworks: Apps SDK, Agent SDKs, LangChain, LangGraph, LlamaIndex, CrewAI, AutoGen.

- Integrations: REST/GraphQL APIs, Webhooks, MCP, OAuth 2.0 and Third-party App Integration.

- Data & Memory: RAG, Embeddings, Vector Databases, Context Management.

- Engineering: Python, TypeScript/Node.js, FastAPI/Express, Docker, AWS/GCP/Azure, CI/CD.

- Quality: Prompt Engineering, Agent Evaluation, Observability, Guardrails.

Skills

Apply

See also

AI Engineering jobs by country — openings, pay and top skills →