Tech Lead (TypeScript)
Company
Orcrist is building a next generation data intelligence platform using cutting-edge technologies. We're handling petabyte-scale data with sub-second queries. Our product (OIP) is a Kubernetes‑based platform delivered as B2B SaaS or as a self‑hosted/on‑prem solution, including air‑gapped deployments.
Role
Lead the team that builds ML‑powered enrichment and search/insight features across our software platform. Think entities and relationships, records browsing, labeling/graphs/profiles, and audio/chat/files (transcription, translation, OCR/NER, summarization, geocoding). You’ll stay hands‑on (ca. 50%), unify batch + streaming pipelines, define typed contracts between our TypeScript services and Python model services, and run inference with clear SLOs and cost controls across SaaS and on‑prem/air‑gapped environments.
What you'll do
-
Own and evolve Insights surfaces: Shaping APIs, data models, and UX of entity Insights, records browser, labeling/graph/profile, and audio/chat/file insights.
-
Integrate ML the right way: Define typed, versioned contracts and rollout gates between TypeScript (Next.js/Node) and Python model services; use feature flags, canaries, and safe rollbacks.
-
Raise search & indexing quality: Manage OpenSearch schemas and analyzers, design reindex/backfill strategies, and run relevance experiments/A‑B tests.
-
Build evaluation & feedback loops: Create offline/online evaluation harnesses (precision/recall, F1, WER/CER, latency/cost SLOs) and integrate human‑in‑the‑loop corrections.
-
Operate streaming + batch enrichment: Design idempotent, backpressure‑aware pipelines (Kafka/Temporal) that support both backfills and real‑time updates.
-
Make it observable and reliable: Instrument traces/metrics/logs end‑to‑end; own SLOs/error budgets and incident response for the Insights surfaces.
-
Manage cost/performance: Reduce inference spend via batching, caching, quantization choices, and right‑sizing GPU/CPU resources; publish and track budgets.
-
Partner across teams: Work with Research on model selection/release, and with Foundation/Platform on data contracts, search/index hygiene, and deployment.
-
Keep data safe: Ensure provenance, auditability, and compliant handling of PII/classified data across SaaS and air‑gapped installs.
About You