Senior Artificial Intelligence/Machine Learning Engineer

About this role

Ciklum is looking for a Senior Artificial Intelligence/Machine Learning Engineer to join our team full-time in India.
We are a custom product engineering company that supports both multinational organizations and scaling startups to solve their most complex business challenges. With a global team of over 4,000 highly skilled developers, consultants, analysts and product owners, we engineer technology that redefines industries and shapes the way people live.
About the role:
As a Senior Artificial Intelligence/Machine Learning Engineer, become a part of a cross-functional development team engineering experiences of tomorrow.
Responsibilities:

• Harness Engineering — Core

• Harness engineering — Excellent understanding of designing agent harnesses: the control loop, tool and skill surface, context and state management, and human-in-the-loop checkpoints around an LLM

• AWS Step Functions with ASL — Author Amazon States Language state machines to orchestrate agent steps, branching, retries, parallelism and human-approval gates (e.g. task-token waits); invoke AgentCore runtimes and Lambda from workflows

• Deterministic vs LLM control — Design workflows where policy gates and rules decide routing and the LLM only proposes, with explicit exception, timeout and loop-guard paths

• Amazon Bedrock AgentCore — Host, operate and scale agent workloads on AgentCore Runtime; understand lifecycle, scaling and invocation patterns, and expose tools via AgentCore Gateway

• AgentCore Observability — Hands-on tracing, metrics and logging of agent runs (AgentCore, CloudWatch, Datadog) with a per-step audit trail to debug and monitor production workflows

• AgentCore Evaluation — Hands-on design and running of evaluations (AgentCore Evaluations, golden-dataset regression tests) to measure agent quality before and after each release

• Guardrails — Hands-on configuration of Amazon Bedrock Guardrails and policy controls for safety, prompt-injection defence, PII and hallucination mitigation, with human-in-the-loop review where required

• Agentic AI with Claude

• Claude Code / Cowork — Experience designing, deploying and scaling AI-powered production workflows using Claude Code and/or Cowork

• LLM API fundamentals — Tool use, function calling, streaming and context management

• Skill building — Design and publish reusable skills consumable by agent workflows

• MCP (Model Context Protocol) — Build, host and consume MCP servers (e.g. FastMCP on AgentCore); expose and integrate tools for agent use, with agents reaching data only through MCP/API, never directly

• Multi-agent coordination — Task delegation and result aggregation across agents (A2A)

• RAG pipelines — Vector databases, chunking, embeddings and retrieval

• Prompt engineering — System prompts, evaluation and iteration

• Front End
• AG-UI — Build agent-facing user interfaces on the AG-UI protocol, connecting the front end to running agents for streaming responses, tool-call visibility, shared state and human approval steps

• Back End

• Python 3.10+ — async/await, type hints

• FastAPI or equivalent Python web framework

• Basic SQL — query, filter, join; awareness of vector DBs (pgvector, Pinecone, OpenSearch)

• NoSQL — Working knowledge of document/key-value stores (e.g. Amazon DynamoDB): data modelling, access patterns, and using them for agent run state and audit trails

• REST API design, JWT / OAuth2

• Workflow Automation

• Decompose manual business processes into agent workflows

• Document processing — PDF, Excel, unstructured data extraction and classification

• AI Engineer

• OCR — Extract text, tables and layout from scanned and image-based documents (e.g. AWS Textract), and feed the output into agent workflows

• Enterprise system integration — SharePoint, email, and external systems via APIs and MCP

• Infrastructure

• AWS services — Lambda, ECS, RDS, S3, IAM — support agent deployment and data storage needs

• Terraform / Infrastructure as Code — Provision and promote AgentCore, Step Functions, Lambda and supporting AWS resources across environments

• Docker and containerisation

• GitHub Actions CI/CD

Requirements:

• Proven experience in software engineering, with hands-on AI engineering shipped to production (years of experience are not the criterion; depth of AI delivery is)

• At least one production project involving LLM agents, skill building, or RAG pipelines — ideally on AWS Bedrock

• AgentCore and/or Step Functions

• Direct experience working with enterprise customers to design, deploy and scale AI-powered production workflows using Claude Code / Cowork

• Track record of owning and delivering solutions end-to-end with minimal guidance

• Experience working directly with non-technical stakeholders — gathering requirements, presenting solutions

• Consulting or client-facing background is an advantage

Desirable:

• Kubernetes

• TypeScript

• Fine-tuning and model evaluation

• CRE domain knowledge

What’s in it for you?

• Strong community: Work alongside top professionals in a friendly, open-door environment

• Growth focus: Take on large-scale projects with a global impact and expand your expertise

• Tailored learning: Boost your skills with internal events (meetups, conferences, workshops), Udemy access, language courses, and company-paid certifications

• Endless opportunities: Explore diverse domains through internal mobility, finding the best fit to gain hands-on experience with cutting-edge technologies

• Care: We’ve got you covered with company-paid medical insurance, mental health support, and financial & legal consultations

About us:
At Ciklum, we are always exploring innovations, empowering each other to achieve more, and engineering solutions that matter. With us, you’ll work with cutting-edge technologies, contribute to impactful projects, and be part of a One Team culture that values collaboration and progress.
India is a strategic innovation hub for Ciklum, with growing teams in Chennai and…