AI Software Engineer

ExTrac

. Build and improve agent-loop components including context assembly, tool selection, and sub-agent orchestration .

Posted 9/30/2026full-timeLondon • United KingdomJuniorMid-Level💰 £75,000 - £80,000 per yearWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in building and improving agent-loop components, including context assembly and tool selection, while effectively managing end-to-end feature ownership and system design. Proficient in Python and familiar with Go, with a strong understanding of distributed systems and cloud infrastructure.

Highest-signal resume keywords

Python ProficiencyEnd-to-End Feature OwnershipDistributed Systems KnowledgeCloud Infrastructure ExperienceAsynchronous Code Writing

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

Software EngineeringProduction-Grade ServicesTest CoverageAgentic Systems ExposureRetrieval SystemsLarge-Scale Vector Database PerformanceGraph AnalysisInfrastructure as CodeStreaming PipelinesMulti-Tenant Systems

Soft Skills

Clear Written CommunicationDocumentation SkillsCuriosity Across Technologies

Tools & Technologies

GoElasticCI/CD PipelinesTrunk-Based DeploymentVersioned APIs

Certifications & Qualifications

UK Security Clearance Eligibility

Industry Keywords

NATO Member State NationalityCompliance-Constrained EnvironmentsFedRAMP

Tech Stack

Tools & technologies

AssemblyCloudDistributed SystemsPythonGo

About the role

Key responsibilities & impact

Build and improve agent-loop components including context assembly, tool selection, and sub-agent orchestration
Build analytical AI features from network construction through analyst-facing summaries
Run experiments against live analyst traffic behind feature flags
Help define what constitutes an improvement and evaluate evidence
Use embeddings for retrieval, network construction, and community detection
Work within a model-agnostic architecture, swapping models and embeddings based on ML evaluations
Clarify requirements and ship well-defined problems to production
Own features and components end to end and contribute to system design and architecture
Build and maintain versioned APIs for internal teams and customers
Design schemas, indexing strategies, and access patterns across relational, document, and vector stores
Build and operate supporting Python and Go services
Improve agent workflows for latency, cost, and reliability
Build and operate long-running streaming pipelines, including caching and recovery
Work directly with product analysts to turn practical usage into system changes
Iterate on product requests while identifying conflicts with longer-term capability work
Partner with the ML team to productionize agentic approaches
Integrate and serve trained models and contribute to production infrastructure for evaluation frameworks

Requirements

What you’ll need

Applicants must be eligible to obtain UK security clearance
Applicants must be nationals of a NATO member state, Australia, or New Zealand
2+ years of professional software engineering experience
Demonstrated ability to build and ship production-grade services with solid test coverage
Experience owning features or components end to end within a larger system
Proficiency building services in Python
Working knowledge of Go or ability to pick it up quickly
Exposure to agentic systems, LLM applications, or retrieval running in production, or clear evidence of ability to learn quickly
Good understanding of distributed systems and databases
Ability to write asynchronous code that performs under load
Experience with cloud infrastructure and CI/CD pipelines
Comfortable working with trunk-based deployment
Ability to diagnose and fix issues such as memory leaks through profiling
Ability to scope technical approaches from well-defined requirements
Clear written communication and documentation skills
Breadth and curiosity across deployment pipelines, database behaviour, security, AI guardrails, and analytical-output validation
Ability to pick up unfamiliar tools quickly
Desirable: retrieval systems and large-scale vector database performance, including Elastic
Desirable: graph or network analysis at scale
Desirable: multilingual retrieval or analysis
Desirable: infrastructure as code, streaming pipelines, and search infrastructure
Desirable: multi-tenant systems with hard data-isolation requirements
Desirable: compliance-constrained environments, including FedRAMP

Benefits

Comp & perks

Competitive salary based on skills and experience
Private Medical Health Insurance
Enhanced pension contributions
Enhanced parental leave
Workplace nursery scheme
£500/year education budget
More expensive educational items, such as conferences, covered with manager approval
33 days of leave across the year inclusive of bank holidays
Flexible working
Team typically in the central London office two days a week, with option to come in up to five days