Website:
synaptyx.ai
Job details:
The Role
We are looking for an AI Associate with 1–2 years of experience who wants to build AI systems that actually make it into users' hands. You’ll work across our core AI solution suite Lattice, with a particular focus on SynSights, our AI powered conversational analytics platform.
Your job is to help us keep making that system better. This is not a role where you spend six months building isolated proof-of-concepts that never leave a notebook. You’ll be working on an existing production codebase: developing new features, improving agent workflows, building APIs, working through data problems, fixing things that don't behave the way they should, and taking features from an idea through to something deployed and usable. We don't expect you to know everything on Day 1. We do expect you to be able to figure things out. No defined playbook, I'm afraid. If shipping working solutions every fortnight, and explaining what you've built to Senior Leadership as fluently as to a fellow engineer is the kind of work that gets you out of bed in the morning, we'd very much like to hear from you.
What You'll Own
- AI & Agentic Development: Build and improve AI workflows using Python, LangChain, and LangGraph. You’ll work on agent orchestration, tool calling, MCP, structured LLM interactions, context handling, memory, retrieval, validation, and the guardrails required to turn an AI workflow into a reliable product.
- SynSights Development: Take ownership of features within SynSights, our conversational analytics platform. That means understanding an existing system before changing it, adding capabilities without breaking what already works, debugging across components, and continuously improving the quality of the answers and analysis the platform produces.
- Data Modelling: Understand unfamiliar client datasets, work out what the tables and columns actually represent, establish relationships between datasets, map business concepts to the underlying data, and structure that understanding so both the application and its AI agents can use it reliably. Work confidently with PostgreSQL: writing and debugging SQL, designing and evolving schemas, understanding joins and relationships, working with large analytical datasets, and thinking about query correctness and performance rather than treating the database as somewhere data happens to live.
- Semantic understanding & context: Help translate raw client data into something an AI system can reason over. You’ll work with schema relationships, metadata, data mappings, metrics, and business definitions so that the platform understands the difference between a column name and what that column actually means.
- API Development: Build and maintain production APIs using FastAPI, including request/response models, authentication-aware endpoints, database interactions, service integrations, error handling, and the backend contracts consumed by the frontend.
- Cloud Infra & Deployment (AWS / Azure): Architect and implement solutions across cloud platforms (AWS, Azure, etc.), ensuring flexibility, performance, scalability, and cost-efficiency. Apply Kubernetes and containerisation best practices to deploy and orchestrate cloud-native and GenAI/ AI-ML workloads, with security frameworks across APIs, identity, and access management (SSO, RBAC) built in from the start.
- CI/CD and DevOps: GitHub Actions, pipelines, and the practices that keep deployments consistent and the team moving.
- Frontend Integration: You don't need to be a frontend specialist, but you do need enough exposure to modern frontend development to trace functionality end to end, understand API integration, make smaller UI changes where needed, and work effectively across the boundary between backend and frontend.
- End-to-End Ownership: Take a task from requirement to implementation to deployment. Ask questions when the requirement genuinely needs clarification; make sensible assumptions when it doesn't. "My part was done" isn't particularly useful if the feature still doesn't work.
What You Bring
- 1–2 years of experience in AI engineering, software engineering, data engineering and experience working with large analytical datasets, query optimisation, partitioning, rollups, or data-ingestion pipelines.
- Strong hands-on Python fundamentals. You should be comfortable reading an unfamiliar Python codebase, debugging it, extending it, and structuring your own code properly.
- Practical experience with LangChain and LangGraph, with an understanding of how LLM applications move beyond a single prompt into stateful workflows, agents, tools, routing, memory, and multi-step execution.
- Working knowledge of vector search and retrieval systems, particularly OpenSearch or equivalent vector databases/search engines.
- Knowledge of MCP / FastMCP and how tool-based communication fits into modern agentic systems. You should understand how to define and expose tools, connect tools to agents, and build reliable agent-to-tool and service-to-agent communication.
- Experience building APIs with FastAPI or a comparable Python API framework.
- Strong working knowledge of SQL and PostgreSQL. Joins, aggregations, CTEs, schemas, relationships, query debugging, and database interactions should be familiar territory.
- A genuinely good understanding of data. Given a set of unfamiliar tables, you should be able to investigate them, understand how they relate, identify keys and useful fields, map business requirements onto them, and reason about whether the resulting analysis actually makes sense.
- Experience with data modelling and data mapping, particularly translating messy source data into structures that applications and analytical systems can reliably consume.
- Working knowledge of AWS, particularly services such as RDS, S3, Lambda, EventBridge, ECR, ECS/Fargate, and EKS. We don't expect you to have operated every service independently in production, but cloud architecture and deployment shouldn't be completely new territory. Familiarity with Docker, containers, APIs, Git, and modern software-development practices. Solid grasp of cloud-native design: cost optimisation, security, and performance tuning, and the trade-offs between all three.
- Experience working on conversational analytical solutions, or systems where an AI-generated answer must remain grounded in real data.
Bonus Points
- Industry experience in Finance, Telecom, Retail, or CPG.
- Practical exposure to machine learning on large datasets, including preparing training data, training and evaluating models, and working with forecasting and prediction problems.
- Hands-on work building AI agents, agentic orchestration, or LLM/SLM-based workflow automation: MCPs, microservices, token and cost optimisation, model selection, prompt engineering, and context window trade-offs.
- Data engineering or analytics background: building analytics-ready datasets, cataloguing, metadata management, and governance frameworks. Exposure to predictive analytics or applied forecasting.
The Mindset We're Hiring For
A figure-it-out mindset. Someone comfortable with ambiguity, proactive in problem solving, and genuinely curious about what's next. You ask the right questions, make reasonable assumptions, ship something, and iterate. You own what you build and can defend it confidently in a client room, even when the client has opinions.
Why SynaptyX
You will not be working on anything that lives in a PowerPoint. Our Solutions are in client hands and being improved every sprint. Your contributions are visible and attributable. The team is small, senior, and low-ego, built on Big 4 advisory depth and serious AI engineering. You will have the authority that matches the accountability. We move fast. Terrifyingly fast, by most corporate standards.
✅ Do: tell us about the hardest technical trade-off you've made under delivery pressure, and what you'd do differently.
❌ Don't: send "passionate about cloud and AI" with no evidence. We'll assume you mean passionate about the phrase.
Click on Apply to know more.
