About Helio Logistics
Helio Logistics operates an orbital transportation fleet that moves customer payloads from LEO insertion to their final operational orbits — predominantly GEO and MEO. Founded in 2022, the company runs 12 electric-propulsion tug spacecraft and a small mission operations team based in Houston and Toulouse.
Helio's unit economics depend on three things: propellant per delivery, throughput per asset per quarter, and time-to-revenue on each new tug. Their commercial customers want firm delivery dates and the ability to track payloads in flight.
The challenge
Through their first 8 missions, Helio's planning workflow was a hand-rolled toolchain: STK for trajectory primitives, custom Python for the continuous-thrust optimizer, a Notion page for the run-of-show, and a Slack channel for approvals. It worked — barely — at 2-3 assets. At 12 assets and growing, the math stopped working.
We were burning two engineering days on every transfer plan, and we still had two operators on console any time we commanded a maneuver. We knew we were two missions away from missing a customer SLA. [Customer Name] · VP Engineering · Helio Logistics
The specific things that had to change:
- Planning time per transfer. 2 engineering days was the floor with the existing toolchain. Helio's growth model needed it under 2 hours.
- Propellant per transfer. The internal solver was good but not great. Internal benchmarking suggested 8-12% headroom.
- Approval discipline. Slack-based "two thumbs up" wasn't going to pass the next assurance review.
- Customer-facing ETAs. Customers wanted live delivery confidence intervals to write into their downstream contracts. The existing system couldn't expose them.
Approach
Helio evaluated LTL.ai against two alternatives: (1) hiring 4 additional flight software engineers to harden the in-house toolchain, and (2) layering on a legacy ground-software vendor's continuous-thrust add-on. The TACK-based path was selected after a three-day pilot on a real GEO insertion mission.
Phase 1: Pilot (3 days)
LTL.ai mission engineers worked alongside Helio's ops team to run TACK against a live GEO insertion plan for asset TUG-04. The pilot delivered a 9.2% propellant savings on a single transfer and a validated command sequence that Helio commanded into production.
Phase 2: Cutover (14 days)
Twelve assets onboarded over two sprints. Helio's policy file was translated from the existing checklists into policy/transfer.yml, reviewed in PR, and deployed alongside the production environment. Two of Helio's three FD tools were retained for first-quarter cross-checks — both retired after 11 weeks.
Phase 3: Operational
By week 5, every plan was being composed in TACK. By week 9, ChatOps notifications replaced the legacy email-and-Slack workflow. By week 14, customer-facing ETAs were live for two pilot customers via the LTL.ai delivery-confidence API.
Results — first 18 weeks operational
Helio's first quarter on LTL.ai covered nine customer deliveries across LEO, MEO, and GEO insertions. Every plan was solved, simulated, and approved on TACK. Every command was signed and uplinked through the LTL.ai ground-network bridge (KSAT + AWS Ground for primary, Viasat RT for the GEO deliveries).
The numbers speak, but the part that surprised us was the audit posture. Our last assurance review took the auditor two days because they couldn't reconstruct who approved what. The current review wrapped in three hours. [Customer Name] · Mission Director · Helio Logistics
- 9 of 9 customer payloads delivered on or before contractual date
- −11% propellant per transfer averaged across the nine missions vs. the prior toolchain baseline
- 0 command-related incidents across 18 weeks operational
- 2× throughput per asset per quarter vs. forecasted from the in-house path
- 3 hours for the most recent assurance review vs. 2 days previously
What's next
Helio is currently in pilot with the LTL.ai conjunction-response module ahead of its 2026-Q4 expansion into a higher-traffic LEO orbit. They've also been an active contributor to the customer council driving the cooperative-autonomy primitives slated for 2027.
A note on this page: the customer name, quotes, and exact metrics are placeholders until we publish a co-authored case study with a real customer. Structure, format, and framing reflect how the final artifact will read.