Senior / Principal Applied AI Engineer (UK / Europe, Remote)

GoCharting INC.; Chennai District; Remote; INR 13,258,000 - 19,887,000; Full time

Job summary

GoCharting is seeking an experienced software engineer to lead AI initiatives, building agentic workflows and production-grade LLM systems. You'll own end-to-end implementation, from agent loops to evaluation, working with a lean team and reporting to leadership.

This fully remote role requires overlap with EU/UK hours and a strong security-first approach. The position emphasizes hands-on coding in Python and TypeScript, integration with existing tools, and a path to scalable product

Qualifications

• 8+ years of production software engineering experience with APIs, integrations, and distributed systems.
• 2–3+ years building and shipping production LLM/agent systems from scratch against provider APIs.
• Hands-on, fluent in Python and TypeScript; write and own production code.
• Strong evaluation and observability discipline: golden datasets, eval harnesses, tracing, latency/cost guardrails.
• Deep LLM-API fluency with Anthropic/OpenAI/open-weight models, MCP and agent frameworks.
• Cloud delivery experience: AWS and/or DigitalOcean, Docker, CI/CD, SQL/Postgres.
• Sharp buy-vs-build judgment and security/data governance maturity.
• High autonomy in a lean team; EU/UK timezone overlap with CET hours.

Responsibilities

• Build AI agents and agentic workflows from scratch, including tool-calling and orchestrations.
• Ship internal AI automations that remove real toil across engineering, support, and growth tools.
• Develop in-product AI capabilities for trading terminal and developer experience in collaboration with Product and Core Platform Engineering.
• Own the buy-vs-build decision, integrate platforms when beneficial, and build when differentiating.
• Put evaluation, observability, and cost guardrails on everything; establish golden datasets and tracing.
• Operate within security and AI governance standards (SOC 2 / ISO and approved-tooling allowlist).

Skills

Production software engineering

LLM/Agent systems

Python

TypeScript

Evaluation & observability

Cloud & deployment

Autonomy

Security & governance

EU/UK timezone overlap

Tools

AWS

Docker

CI/CD

PostgreSQL

Job description

GoCharting is a web-native charting and trading workflow platform with institutional-grade orderflow and DOM tools - footprint/cluster charts, volume/market profile, cumulative delta, imbalance charts, DOM trading. We are the only platform that delivers these capabilities fully on the web. We are differentiated as the leader in the professional orderflow niche. PE-acquired and funded in 2025, profitable, 3M+ registered traders globally, with a mandate to scale 10x over 4 years.

The Role

We want AI to be a genuine force multiplier across GoCharting - first inside the company, then inside the product. This is the person who builds that: AI agents, agentic workflows, and the pipelines and harnesses that wire large language models into our real systems and, over time, into the product 3M+ traders use.

You will build from scratch when the problem demands it and integrate the best existing platform when it doesn’t - and you’ll know the difference. Early on the weight is internal: removing real toil across engineering, support, and growth by wiring LLMs into the tools we already run and shipping agents that actually work in production. As that muscle matures, the weight shifts toward the product - bespoke, build-heavy AI capabilities in the trading terminal and developer experience, built hand-in-hand with Product and Core Platform Engineering.

This is a hands-on engineering seat, not an advisory one. You write and own production code, you put evaluation and observability on everything you ship, and you run autonomously in a lean team. You report to the COO and partner closely with the CTO, the Head of Product, and Head of IT & Security. In year one, success looks like a set of AI systems live in production - measured, reliable, and paying for themselves in time saved or product value created.

What you'll do

• Build AI agents and agentic workflows from scratch - agent loops, tool-calling, structured outputs, planning and state, retries and guardrails - and the orchestration/harness layer on top of existing LLM providers via API (Anthropic, OpenAI, open-weight models, MCP). You write the agent, not just its config.
• Ship internal AI automations that remove real toil - e.g. triaged, automated engineering-workflow and pull-request review across the tools we already run (GitHub, Slack, Intercom, Jira), support deflection, and social/market listening - integrating an off-the-shelf tool where the need is commodity, and building bespoke where it isn’t.
• Build in-product AI capabilities with Product and Core Platform Engineering - bespoke pipelines for the trading terminal and developer experience where no off-the-shelf option fits (this side is build-heavy). You prototype it, then productionize it with the platform team.
• Own the buy-vs-build call - integrate and measure an existing platform first for commodity internal needs; build or replace it internally when it’s differentiating or becomes cost-prohibitive. Ship outcomes, not architecture debates.
• Put evaluation, observability, and cost guardrails on everything - golden datasets, eval harnesses, tracing, fallback chains, latency and spend controls. Nothing ships as an unmeasured demo.
• Operate inside our security and AI governance - SOC 2 / ISO track and an approved-tooling allowlist - handling source code, customer data, and product data responsibly.

What we're looking for

Must-haves:

• 8+ years of production software engineering, deep in APIs, integrations, and systems - REST, webhooks, OAuth, event-driven and distributed systems - with integrations you owned end-to-end (not “called an API once”).
• 2–3+ years building and shipping production LLM/agent systems from scratch against provider APIs - you have personally written the agent loops, orchestration/harness, tool-calling, structured outputs, guardrails and RAG, running for real users. Assembling an off-the-shelf agent builder does not count on its own.
• Hands-on, fluent in Python and TypeScript - you write and own production code. This is a builder seat: not no-code/low-code wiring, not “vibe coding”, not advisory.
• Evaluation and observability discipline - golden datasets, eval harnesses, tracing, cost/latency guardrails, graceful failure. You treat “does it work, measurably?” as the job, not an afterthought.
• Deep LLM-API fluency (Anthropic/OpenAI/open-weight), MCP, agent frameworks - and fluency using AI coding tools (Claude Code, Cursor) to move fast.
• Cloud and delivery: AWS and/or DigitalOcean, Docker, CI/CD, SQL/Postgres.
• Sharp buy-vs-build judgment and security/data-governance maturity - you don’t pipe production or customer data to arbitrary endpoints.
• High autonomy in a lean team - you find the problem, design the solution, and ship it.
• EU/UK timezone, with solid overlap with Central European hours.

Nice-to-haves:

• Go experience - a strong plus for building and productionizing alongside our platform engineering team (not a gate).
• Fintech / trading / market-data domain, or genuine experience as a trading-platform user.
• Open-source contributions to AI tooling (LangChain, LlamaIndex, vLLM, DSPy, or similar).
• Production RAG and vector databases; code-generation, transpilation, or developer-experience tooling experience.

Logistics

Location Europe (EU/UK) - fully remote; must overlap Central European working hours

Employment type Full-Time (via Employer-of-Record in your home country)

Travel Light - occasional team/leadership meetups